Publications
Disclaimer :
These papers are made available as a means to ensure timely dissemination of scholarly and technical work
on a non-commercial basis. Copyright and all rights therein are maintained by the authors or by other copyright holders,
notwithstanding that they have offered their works here electronically. It is understood that all persons copying
this information will adhere to the terms and constraints invoked by each author's copyright. These works may not
be reposted without the explicit permission of the copyright holder.
2012
Coexist: A Hybrid Approach for Content Oriented Publish/Subscribe Systems ,
Jiachen Chen , Mayutan Arumaithurai , Xiaoming Fu , K. K. Ramakrishnan, ACM SIGCOMM Workshop on Information-Centric Networking (ICN 2012), Helsinki, Finland (in conjunction with ACM SIGCOMM 2012),
ACM, August 2012.
iSATS: Leveraging Identity Based Sender Authentication for Spam Mitigation ,
Sufian Hameed , Tobias Kloht , Xiaoming Fu , , International Conference on Security and Cryptography (Secrypt), Rome, Italy,
July 2012.
G-COPSS: A Content Centric Communication Infrastructure for Gaming ,
Jiachen Chen , Mayutan Arumaithurai , Xiaoming Fu , and K. K. Ramakrishnan, The 32nd IEEE International Conference on Distributed Computing Systems (ICDCS 2012), Macau, China,
June 2012.
NCShield: Securing Decentralized, Matrix Factorization-Based Network Coordinate Systems ,
Shining Wu , Yang Chen , Xiaoming Fu , Jun Li, IEEE/ACM International Workshop on Quality of Service (IWQoS 2012), Coimbra, Portugal,
IEEE, June 2012.
CloudGPS: A Scalable and ISP-Friendly Server Selection Scheme in Cloud Computing Environments ,
Cong Ding , Yang Chen , Tianyin Xu , Xiaoming Fu , IEEE/ACM International Workshop on Quality of Service (IWQoS 2012), Coimbra, Portugal,
IEEE, June 2012.
Exploring Regional and Global Population Growth in Online Social Networks (extended abstract) ,
Konglin Zhu , Wenzhong Li , Xiaoming Fu , Praxis der Netzwerkforschung 2012 (PdN 2012), Frankfurt am Main, Germany,
May 2012.
Thumb: A Real-Time Resource Information Sharing Application over Mobile Phones ,
Narisu Tao , Konglin Zhu , Xiaoming Fu , IEEE International Conference on Pervasive Computing and Communication (PerCom) 2012, demo session, Lugano, Switzerland,
March 2012.
Measurement-based Optimization of P2P Networking and Applications ,
Xiaoming Fu , Yang Chen , Guy Leduc, Laurent Mathy, Computer Networks, special issue on easurement-based Optimization of P2P Networking and Applications (editorial), 56(3): 1077-1079,
Elsevier, February 2012.
Annual Report of Computer Networks Group at the University of Göttingen (2011) ,
Xiaoming Fu (ed.),
January 2012.
Read abstract
Founded in April 2007, the Computer Networks Group is devoted to cutting-edge leading research and education programs in the field of network architectures, protocols and applications. This report series intends to review the most relevant and important research results and other achievements and activities performed by the Computer Networks Group. The present report is the fourth issue of this series, and summarizes those of the year of 2011.
PDF [166.9 kB]
Receiving Buffer Adaptation for High-speed Data Transfer ,
Hao Liu, Yaoxue Zhang, Yuezhi Zhou , Xiaoming Fu , Laurence Yang, IEEE Transactions on Computers (accepted),
IEEE, 2012.
2011
Phoenix: A Weight-based Network Coordinate System Using Matrix Factorization ,
Yang Chen , Xiao Wang, Cong Shi, Eng Keong Lua, Xiaoming Fu , Beixing Deng, Xing Li, IEEE Transactions on Network and Service Management, Vol 8, Issue 4,
IEEE, December 2011.
Read abstract
Network coordinate (NC) systems provide a lightweight and scalable way for predicting the distances, i.e., round-trip latencies among Internet hosts. Most existing NC systems embed hosts into a low dimensional Euclidean space. Unfortunately, the persistent occurrence of Triangle Inequality Violation (TIV) on the Internet largely limits the distance prediction accuracy of those NC systems. Some alternative systems aim at handling the persistent TIV, however, they only achieve comparable prediction accuracy with Euclidean distance based NC systems. In this paper, we propose an NC system, so-called Phoenix, which is based on the matrix factorization model. Phoenix introduces a weight to each reference NC and trusts the NCs with higher weight values more than the others. The weight-based mechanism can substantially reduce the impact of the error propagation. Using the representative aggregate data sets and the newly measured dynamic data set collected from the Internet, our simulations show that Phoenix achieves significantly higher prediction accuracy than other NC systems. We also show that Phoenix quickly converges to steady state, performs well under host churn, handles the drift of the NCs successfully by using regularization, and is robust against measurement anomalies. Phoenix achieves a scalable yet accurate end-to-end distances monitoring. In addition, we study how well an NC system can characterize the TIV property on the Internet by introducing two new quantitative metrics, so-called RERPL and AERPL. We show that Phoenix is able to characterize TIV better than other existing NC systems.
Phoenix_TNSM.pdf [433.5 kB]
GEMSTONE: Empowering Decentralized Social Networking with High Data Availability ,
Florian Tegeler , David Koll , and Xiaoming Fu , IEEE GLOBECOM 2011 - Selected Areas in Communications Symposium - Social Networks Track, Houston, TX, USA,
IEEE, December 2011.
Read abstract
Social networking platforms such as Facebook, MySpace, Twitter, and many others have seen an enormous increase in user population and user provided information. However, users are increasingly concerned about identity and data privacy since information is aggregated at single companies. To address this issue researchers have been investigating alternative solutions, where the users’ data such as profile information, comments and messages is stored at user-controlled nodes. Although these solutions provide a plausible means for avoiding privacy leaking in central instances, they raise a new challenge to design a cost-effective storage replica scheme which ensures a high data availability even when some users are offline. In this paper we present Gemstone, a social network platform, where the data replication scheme leverages a learning mechanism based on social relationships, online patterns of peers and user experiences. Our preliminary evaluation shows that compared to related works, it achieves higher data availability while only requiring a smaller number of data replicas.
PDF [434.9 kB]
Scaling Microblogging Services with Divergent Traffic Demands ,
Tianyin Xu , Yang Chen , Lei Jiao , Ben Y. Zhao, Pan Hui, Xiaoming Fu , ACM/IFIP/USENIX 12th International Middleware Conference (Middleware 2011), Lisboa, Portugal, Lecture Notes in Computer Science 7049, pages 20-40,
Springer Verlag, December 2011.
Read abstract
Today’s microblogging services such as Twitter have long outgrown their initial designs as SMS-based social networks. Instead, a massive and steadily-growing user population of more than 100 million is using Twitter for everything from capturing the mood of the country to detecting earthquakes and Internet service failures. It is unsurprising that the traditional centralized client-server architecture has not scaled with user demand, leading to server overload and significant loss of availability.
In this paper, we argue that the divergence in usage models of microblogging services can best be addressed using complementary mechanisms, one that provides reliable messages between friends, and another that delivers events from popular celebrities and media outlets to their millions of followers. We present Cuckoo, a new microblogging system that offloads processing and bandwidth costs away from a small centralized server base while ensuring reliable message delivery. We use a 20-day Twitter availability measurement to guide our design, and trace-driven emulation of 30,000 Twitter users to evaluate our Cuckoo prototype. Compared to a centralized approach, Cuckoo achieves 30-50% server bandwidth savings and 50-60% CPU load reduction, all while guaranteeing reliable message delivery.
cuckoo.pdf [2593.3 kB]
Latency-Aware Data Partitioning for Geo-Replicated Online Social Networks ,
Lei Jiao , Tianyin Xu , Jun Li, Xiaoming Fu , ACM/IFIP/USENIX 12th International Middleware Conference (Middleware 2011), Poster session, Lisboa, Portugal,
December 2011.
Read abstract
No abstract is associated with this entry. Please download the two-page poster directly. Thanks.
TAMER.pdf [156.4 kB]
COPSS: An Efficient Content Oriented Publish/Subscribe System ,
Jiachen Chen , Mayutan Arumaithurai , Lei Jiao , Xiaoming Fu , K. K. Ramakrishnan, ACM/IEEE Symposium on Architectures for Networking and Communications Systems (ANCS 2011), Brooklyn, NY, USA,
October 2011.
Read abstract
Content-Centric Networks (CCN) provide substantial
flexibility for users to obtain information without regard to the source of the information or its current location. Publish/subscribe (pub/sub) systems have gained popularity in society to provide the convenience of removing the temporal dependency of the user having to indicate an interest each time he or she wants to receive a particular piece of related information. Currently, on the Internet, such pub/sub systems have been built on top of an IP-based network with the additional responsibility placed on the end-systems and servers to do the work of getting a piece of information to interested recipients. We propose Content-Oriented Pub/Sub System (COPSS) to achieve an efficient pub/sub capability for CCN. COPSS enhances the heretofore inherently pull-based CCN architectures proposed by integrating a push based multicast capability at the content-centric layer.
We emulate an application that is particularly emblematic
of a pub/sub environment - Twitter - but one where subscribers are interested in content (e.g., identi fied by keywords), rather than tweets from a particular individual. Using trace-driven simulation, we demonstrate that our architecture can achieve a scalable and efficient content centric pub/sub network. The simulator is parameterized using the results of careful microbenchmarking of the open source CCN implementation and of standard IP based forwarding. Our evaluations show that COPSS provides considerable performance improvements in terms of aggregate network load, publisher load and subscriber experience compared to that of a traditional IP infrastructure.
PDF [1102.1 kB]
Building Mega Data Centers from Heterogeneous Containers ,
Dan Li, Mingwei Xu, Hongze Zhao, Xiaoming Fu , IEEE International Conference on Network Protocols (ICNP 2011), Vancouver, Canada,
IEEE, October 2011.
Read abstract
Data center containers are regarded as the basic units to build mega data centers. In practice, heterogeneity exists among data center containers, because of technical innovation and vendor diversity. In this paper, we propose uFix, a scalable, flexible and modularized network architecture to interconnect heterogeneous data center containers. The inter-container connection rule in uFix is designed in such a way that it can flexibly scale to a huge number of servers with stable server/switch hardware settings. uFix allows modularized and fault-tolerant routing by completely decoupling inter-container routing from intra-container routing. We implement a software-based uFix stack on the Linux platform. Simulation and experiment results show that uFix enjoys high network capacity, gracefully handles server/switch failures, and brings light-weight CPU overhead onto data center servers.
PDF [2045.2 kB]
LENS: Leveraging Social Networking and Trust to Prevent Spam Transmission ,
Sufian Hameed , Xiaoming Fu , Pan Hui and Nishanth Sastry, The Second IEEE Workshop on Security & Trust in the Future Internet (FIST'11), Co-located with IEEE ICNP 2011,
Vancouver, BC Canada, October 2011.
Read abstract
In this paper we introduce LENS, a novel spam protection system based on the recipient’s social network, which allows correspondence within the social circle to directly pass to the mailbox and further mitigates spam beyond social circles. The key idea in LENS is to select legitimate and authentic users, called Gatekeepers (GKs), from outside the recipients social circle and within pre-defined social distances. Unless a GK vouches for the emails of potential senders from outside the social
circle of a particular recipient, those e-mails are prevented from transmission. In this way LENS drastically reduces the consumption of Internet bandwidth by spam. Using extensive evaluations, we show that LENS provides each recipient reliable email delivery from a large fraction of the social network. We also evaluate the computational complexity of email processing with LENS deployed on two Mail Servers (MSs) and compared it with the most popular content-based filter i.e SpamAssassin.
LENS proved to be fast in processing emails (around 2-3 orders of magnitude better than SpamAssassin) and scales efficiently with increasing community size and GKs.
PDF [398.1 kB]
G-COPSS: A Content Centric Communication Infrastructure for Gaming Applications ,
Jiachen Chen , Mayutan Arumaithurai , Xiaoming Fu , K. K. Ramakrishnan, 18th IEEE International Workshop on Local and Metropolitan Area Networks (LANMAN 2011), Chapel Hill, NC, USA,
IEEE, October 2011.
Read abstract
With users increasingly focused on an online world,
an emerging challenge for the network infrastructure is the need to support Massively Multiplayer Online Role Playing Games (MMORPG). This is an application domain that is attracting more players than ever before, very often with players distributed over a metropolitan area. Currently, MMORPG are built on an IP infrastructure with the primary responsibility on servers to do the work of disseminating control messages and having to predict/retrieve objects in each player’s view. Limited server resources significantly impair the user’s interactive experience. Modern fast-paced action games that run on a client/server architecture limit the number of players who can interact simultaneously since the server needs to handle the frequent updates and disseminate them. Scale and timeliness are major challenges of such a server-oriented gaming architecture.
We propose Gaming over COPSS (G-COPSS), a communication
infrastructure using a Content-Oriented Pub/Sub System (COPSS) to enable efficient decentralized information dissemination in MMORPG, exploiting the network and the end-systems for player management and information dissemination. We emulate an application that is particularly emblematic of MMORPG – Counter-Strike – but one in which all the players share a hierarchical structured map. Using trace-driven simulation, we demonstrate that G-COPSS can achieve high scalability and
tight timeliness requirements of MMORPG. The simulator is
parameterized using the results of careful microbenchmarking
of the open-source CCN implementation and of standard IP based forwarding. Our evaluations show that G-COPSS provides
considerable performance improvement in terms of aggregate network load and update latency compared to that of a traditional IP server-based infrastructure.
PDF [1355.4 kB]
Low Overhead Dynamic Spectrum Reallocation in Opportunistic Spectrum Access Networks ,
Yu Tian, Wenzhong Li , Zhuo Li, Sanglu Lu, Xiaoming Fu , Proceedings of IEEE 20th International Conference on Computer Communications and Networks (ICCCN 2011),
August 2011.
Read abstract
Opportunistic spectrum access, which allows secondary users opportunistically access unused licensed channels to exploit instantaneous spectrum availability, is a promising approach to achieve efficient spectrum utilization and mitigate spectrum scarcity. To address the challenge of dynamic spectrum access, one of the most important issues is cooperative spectrum reallocation among secondary users to minimize spectrum handoffs. In this paper, we present a low-complexity approach based on conflict graph to optimize spectrum reallocation by local coordination. In order to reduce communication overhead, we propose two heuristic spectrum selection methods named local observation and metric maximum. Experimental results show two benefits of the proposed schemes. On one hand, the coordination approach can dynamically improve the total system throughput (approximately doubled at most). On the other hand, our heuristic strategies decrease the number of spectrum handoff by up to 20% compared with existing strategies.
PDF [186.9 kB]
Geo-Assisted Multicast Inter-Domain Routing (GMIDR) Protocol for MANETs ,
Konglin Zhu , Kelvin (Biao) Zhou, Xiaoming Fu , and Mario Gerla, IEEE International Conference on Communications (ICC) 2011, Kyoto, Japan,
IEEE, June 2011.
Read abstract
Large military ad hoc networks are often characterized by the interconnection of heterogeneous domains. The same trend is emerging in civilian MANETs (e.g., search and rescue, vehicular networks). In these networks it is important to be able to efficiently propagate information across domains in multicast mode (e.g., situation awareness dissemination, commands, streams). Several multicast protocols have been developed for single domain MANET. However, few can be extended to inter-domain operation. In fact, multicast routing across different MANET domains faces the challenges of node motion, topology changes, dynamic gateway election and, possibly, connectivity interruption. To overcome these challenges, especially to achieve routing scalability and at the same time maintains efficient routing, this paper proposes the Geo-assisted Multicast Inter-domain Routing (GMIDR) protocol based on geographical assistance and cluster technology. Intensive simulation results show that the GMIDR protocol is scalable and stable with various numbers of multicast group members, and it outperforms other multicast protocols. Geocast by applying GMIDR shows the flexibility of the protocol.
PDF [177.4 kB]
COPS: An Efficient Content Oriented Publish/Subscribe System ,
Jiachen Chen , Mayutan Arumaithurai , Lei Jiao , Xiaoming Fu , K. K. Ramakrishnan, Technical Report No. IFI-TB-2011-06, Institute of Computer Science, University of Goettingen, Goettingen, Germany,
ISSN 1611-1044, June 2011.
Read abstract
Content-Centric Networks (CCN) provide substantial flexibility for users to obtain information without regard to the source of the information or its current location.
Publish/subscribe (pub/sub) systems have gained popularity in society to provide the convenience of removing the temporal dependency of the user having to indicate an interest each time he or she wants to receive a particular piece of related information. Currently, on the Internet, such pub/sub systems have been built on top of an IP-based
network with the additional responsibility placed on the end-systems and servers to do the work of getting a piece of information to interested recipients. We propose Content-Oriented Pub/Sub system (COPS) to achieve an efficient pub/sub capability for CCN. COPS enhances the heretofore inherently pull-based CCN architectures proposed by integrating push based multicast at the content-centric layer.
We emulate an application that is particularly emblematic of a pub/sub environment---Twitter---but one where subscribers are interested in content (e.g., identified by keywords), rather than tweets from a particular individual. Using trace-driven simulation, we demonstrate that our architecture can achieve a scalable and efficient pub/sub content centric network. The simulator is parameterized using the results of careful microbenchmarking of the open source CCN implementation and of standard IP based forwarding. Our evaluations show that COPS provides considerable performance improvements in terms of aggregate network load, publisher load and subscriber experience compared to that of a traditional IP infrastructure.
main.pdf [1488.3 kB]
NF-TCP: A Network Friendly TCP Variant for Background Delay-Insensitive Applications ,
Mayutan Arumaithurai , Xiaoming Fu , K. K. Ramakrishnan, Proc. 10th IFIP International Conference on Networking (Networking 2011), Valencia, Spain,
Springer-Verlag, May 2011.
Read abstract
Delay-insensitive applications, such as P2P file sharing, generate substantial amounts of traffic and compete with other applications on an equal footing when using TCP. Further, to optimize throughput, such applications typically open multiple connections. This results in unfair and potentially poor service for applications that have stringent performance objectives (including sensitivity to delay and loss). In this paper, we propose NF-TCP, a TCP variant for P2P and similar delay-insensitive applications that can afford to have communication in the “background”. NF-TCP aims to be submissive to delay-sensitive applications under congestion. A major component of NF-TCP is to integrate measurement as an integral component of the congestion control framework. This enables the transport to exploit available bandwidth, so that it can aggressively utilize spare capacity.
We implemented NF-TCP on Linux and ns-2. Our evaluations of the NF-TCP Linux implementation on ns-2 show that NF-TCP outperforms other network friendly approaches (e.g., LEDBAT, TCP-LP and RAPID). NF-TCP achieves high utilization, fair bandwidth allocation among NF-TCP flows and maintains a small average queue. Our evaluations further demonstrate that with NF-TCP, the available bandwidth can be efficiently utilized.
PDF [705.5 kB]
PS-CCN: Achieving an Efficient Publish/Subscribe Capability for Content-Centric Networks ,
Jiachen Chen , Lei Jiao , Mayutan Arumaithurai , Xiaoming Fu , K. K. Ramakrishnan, Technical Report No. IFI-TB-2011-04, Institute of Computer Science, University of Goettingen, Goettingen, Germany,
ISSN 1611-1044, May 2011.
Read abstract
Content-Centric Networks (CCN) provide a substantial degree of flexibility for users and end-systems to obtain information without regard to their location or source. Publish/subscribe (pub/sub) systems have gained popularity in society to provide the same convenience for users. Currently, on the Internet, such pub/sub systems have been built on top of an IP-based network with additional responsibility placed in the end-systems and servers to do the work of dissemination. We propose PS-CCN to achieve an efficient pub/sub capability for CCN. PS-CCN enhances the inherently pull-based CCN by introducing the capability of multicast. We use trace-driven simulations to emulate an application that is particularly emblematic of a pub/sub environment—Twitter—but one where subscribers are interested in content (e.g.,
identified by keywords), rather than tweets from a particular individual. The simulator is tuned using the result of a careful microbenchmarking of the open source CCN implementation and standard IP based forwarding. Our evaluations show that PS-CCN provides considerable performance improvements in terms of aggregate network load, publisher load and subscriber experience compared to that of the traditional IP infrastructure.
ccnTechreportMay.pdf [1562.5 kB]
Exploiting Graphics Processors for High-performance IP Lookup in Software Routers ,
Jin Zhao , Xinya Zhang, Xin Wang, Yiaodong Deng, and Xiaoming Fu , IEEE INFOCOM 2011, Mini-conference, Shanghai, China,
April 2011.
Read abstract
As the physical link speeds grow and the size of routing table continues to increase, IP address lookup has been a challenging problem at routers. There have been growing demands in achieving high-performance IP lookup cost effectively. Existing approaches typically resort to specialized hardwares, such as TCAM. While these approaches can take advantage of hardware parallelism to achieve high-performance IP lookup, they also have the disadvantage of high cost. This paper investigates a new way to build a cost-effective IP lookup scheme using graphics processor units (GPU). Our contribution here is to design a practical architecture for high-performance IP lookup engine with GPU, and to develop efficient algorithms for routing prefix update operations such as deletion, insertion, and modification. Leveraging GPU’s many-core parallelism, the proposed schemes addressed the challenges in designing IP lookup at GPU-based software routers. Our experimental results on real-world route traces show promising gains in IP lookup and update operations.
PDF [349.1 kB]
Exploring User Social Behaviors in Mobile Social Applications ,
Konglin Zhu , Pan Hui, Yang Chen , Xiaoming Fu , Wenzhong Li , 4th ACM Workshop on Social Network Systems (SNS 2011), in conjunction with ACM EuroSys 2011, Salzburg, Austria,
April 2011.
Read abstract
Mobile social applications are popular as the proliferating of mobile devices. Understanding user social behaviors is important to improve mobile social applications and enhance its quality of service. However, there is still lack of data for real deployment mobile social application on data analysis of human interaction and social behaviors in mobile social networks.
In this paper, we introduce the experiment methodology of deploying the Goose software in two campuses located in Germany and China respectively. Goose is a mobile social network application allows microblogging, message sending. With the help of volunteers, we collect user interaction data in the duration of 15 days. Based on the collected data, our observation reveals the following aspects of user interactions and their influences. First, user overall activities approximately match user daily life work pattern
with a slightly longer time duration and periodically appearance. Second, user encounters in mobile social network follow the heavy tail distribution in small social communities, and user interactions follow the Pareto principle, where about 20% of users make close connections to the other users. Third, communication path between a pair of mobile nodes is mostly within 6 hops, and information diffusion using an epidemic strategy demonstrates that the informed population reaches to 50% in a short term and approaches to 80% in a long term.
PDF [185.3 kB]
LENS: LEveraging Social Networking and trust to prevent Spam transmission ,
Sufian Hameed , Pan Hui, Xiaoming Fu , Nishanth Sastry, 5th ACM EuroSys Doctoral Workshop (EuroDW 2011), Salzburg, Austria,
April 2011.
NSIS Protocol Operation in Mobile Environments ,
Takako Sanda, Xiaoming Fu , Seong-Ho Jeong, Jukka Manner, and Hannes Tschofenig , Internet Engineering Task Force, RFC 5980, Next Steps in Signaling (NSIS) Working Group,
ISSN: 2070-1721, March 2011.
Read abstract
Mobility of an IP-based node affects routing paths, and as a result, can have a significant effect on the protocol operation and state management. This document discusses the effects mobility can cause to the Next Steps in Signaling (NSIS) protocol suite, and shows how the NSIS protocols operate in different scenarios with mobility management protocols.
PDF [68.0 kB]
LENS : LEveraging social Networking and trust to prevent Spam transmission ,
Sufian Hameed , Pan Hui, Xiaoming Fu and Nishanth Sastry, Technical Report No. IFI-TB-2011-02,
Institute of Computer Science, University of Göttingen, Germany, ISSN 1611-1044, February 2011.
Read abstract
We introduce LENS, a novel spam protection system which leverages the social network of the recipient. In LENS, previous correspondents with a recipient are allowed to send emails directly. To enable new senders to send emails, each recipient independently picks a small number of Gatekeeper (GK) nodes dispersed in the social network. Each GK is authorized to generate voucher, and new senders are required to obtain a voucher (to communicate with the recipient) from a GK in their social neighbourhood. Recipients recover from compromised GKs simply by selecting replacements and revoking vouchers. Using extensive evaluations, we show that LENS provides each recipient reliable email delivery from a large fraction (upto 55% of entire userbase) of the social network. LENS also proved to be effective in accepting all the legitimate inbound emails from the real email traces. LENS imposes zero overhead for the common case of frequent and familiar senders, and remains lightweight for the general case:
Our prototype implementation of LENS in Postfix/MailAvenger show that LENS consumes up to 75% less CPU and 9% less memory as traditional solutions like SpamAssassin.
LENS-ISSN 1611-1044 [670.5 kB]
General Internet Signaling Transport (GIST) over Stream Control Transmission Protocol (SCTP) and Datagram Transport Layer Security (DTLS) ,
Xiaoming Fu , Christian Dickmann , and Jon Crowcroft, RFC 6084, Internet Engineering Task Force, Next Steps in Signaling (NSIS) Working Group,
January 2011.
Read abstract
The General Internet Signaling Transport (GIST) protocol currently uses TCP or Transport Layer Security (TLS) over TCP for Connection mode operation. This document describes the usage of GIST over the Stream Control Transmission Protocol (SCTP) and Datagram Transport Layer Security (DTLS).
RFC6084.pdf [23.5 kB]
Annual Report of Computer Networks Group at the University of Göttingen (2010) ,
Xiaoming Fu (ed.) ,
January 2011.
Read abstract
This document reports the major activities performed and results achieved by the Computer Networks Group at the University of Göttingen throughout the calendar year 2010.
ANNUAL_REPORT_2010.pdf [106.4 kB]
Cuckoo: Scaling Microblogging Services with Divergent Traffic Demands ,
Tianyin Xu , Yang Chen , Lei Jiao , Ben Y. Zhao, Pan Hui, Xiaoming Fu , Technical Report No. IFI-TB-2011-01, Institute of Computer Science, University of Goettingen, Goettingen, Germany, ISSN 1611-1044,
January 2011.
Read abstract
Today's microblogging services such as Twitter have long outgrown their initial designs as SMS-based social networks. Instead, a massive and steadily-growing user population of more than 100 million is using Twitter for everything from capturing the mood of the country to detecting earthquakes and Internet service failures. It is unsurprising then, that the traditional centralized client-server architecture has not scaled with user demand, leading to server overload and significant loss of availability.
In this paper, we argue that the divergence in usage models of microblogging services can best be addressed using complementary mechanisms, one that provides reliable messages between friends, and another that delivers events from popular celebrities and media outlets to their millions of followers. We present Cuckoo, a new microblogging system that offloads processing and bandwidth costs away from a small centralized server base while ensuring reliable message delivery. We use a 20-day Twitter availability measurement to guide our design, and trace-driven emulation of 30,000 Twitter users to evaluate our Cuckoo prototype. Compared to a centralized approach, Cuckoo achieves 30-50% server bandwidth savings and 50-60% CPU load reduction, all while guaranteeing reliable message delivery.
Technical Report.pdf [4259.3 kB]
2010
An Experimental Analysis of Joost Peer-to-Peer VoD Service ,
Jun Lei , Lei Shi , Xiaoming Fu , Peer-to-Peer Networking and Applications, Volume 3, Number 4, Pages 351-362,
Springer Verlag, ISSN 1936-6442, December 2010.
Read abstract
Despite strong interest in peer-to-peer (P2P) Video-on-Demand (VoD) services, existing studies mostly focus on peer-to-peer or overlay protocol design based on simulations under various topological constraints. We believe experimental studies on a real-life P2P VoD system will provide valuable information to ISPs, network administrators, and content owners. In this paper we present a comprehensive analytical and experimental study on Joost, one of the first commercial P2P VoD systems used for distributing various forms of video over the Internet. Our extensive experiments prove that Joost is a server-assisted peer-to-peer VoD system. With several envisioned typical scenarios we have further investigated the peer management in terms of time pattern, bandwidth consumption and locality considerations. Our major findings include: (1) the current Joost system is capable of providing high-quality VoD service through the use of an overlay network deployed with a set of centralized content servers; (2) inter-continental links are often used regardless of the number of local users, which may pose a high burden on the network providers; (3) easily reachable, high-capacity nodes are selected as main relaying nodes, similar to super nodes in Skype, to facilitate the traversal of symmetric NATs and firewalls. We also provide insights on the potential ways to construct more efficient P2P VoD systems (e.g. considering topological locality-awareness, using adaptive/layered video).
PDF.pdf [601.9 kB]
Taming the Triangle Inequality Violations with Network Coordinate System on Real Internet ,
Yibo Zhu, Yang Chen , Zengbin Zhang, Xiaoming Fu , Dan Li, Beixing Deng, Xing Li, ACM Workshop on Re-Architecting the Internet (ReArch 2010), co-located with ACM CoNEXT 2010, Philadelphia, USA,
ACM, December 2010.
Read abstract
Network Coordinate (NC) systems are efficient in scalable Internet latency estimation. While most of the focus has been put on how to distort Triangle Inequality Violation (TIV) in metric spaces to relieve the inaccuracy caused by it, TIV is a persistently and widely existing phenomenon on the Internet and thus should be embraced by future NC systems rather than being eliminated. Besides high accuracy, such an NC system can also provide the benefit of reducing the data transmission time by use of proper relay routes. With that in mind, we design an NC system with a hierarchical architecture, which is motivated by the natural idea of partitioning the three TIV links into different autonomous NC systems, in order to make as many as TIVs inherently embeddable in metric space. We implement and deploy our work, named Toread, on real Internet. Evaluation results show that Toread 's metric space can well characterize more than 60% TIVs, thus Toread is highly accurate (0.54 in Toread versus 1.06 in Pyxida at 90th percentile Relative Error) and effective in searching detour paths (succeeds in 58.2% cases).
Toread_ReArch10.pdf [372.8 kB]
General Internet Signaling Transport (GIST) State Machine ,
Tseno Tsenov, Hannes Tschofenig , Xiaoming Fu , Cedric Aoun, and Elwyn Davies, Internet Engineering Task Force, Request for Comment (RFC) 5972, Next Steps in Signaling (NSIS) Working Group,
October 2010.
Read abstract
This document describes state machines for the General Internet Signaling Transport (GIST). The states of GIST nodes for a given flow and their transitions are presented in order to illustrate how GIST may be implemented.
PDF [34.5 kB]
Twittering by Cuckoo -- Decentralized and Socio-Aware Online Microblogging Services ,
Tianyin Xu , Yang Chen , Xiaoming Fu , Pan Hui, ACM SIGCOMM Computer Communication Review, Volume 40, Issue 4, Pages 473-474,
ACM, October 2010.
Read abstract
Online microblogging services, as exemplified by Twitter, have become immensely popular during the latest years. However, current microblogging systems severely suffer from performance bottlenecks and malicious attacks due to the centralized architecture. Thus, centralized microblogging systems may threaten the scalability, reliability as well as availability of the offered services, not to mention
the high operational and maintenance cost.
This demo presents a decentralized, socio-aware microblogging system named Cuckoo. The key aspects of Cuckoo’s design is to take advantage of the inherent social relations while leveraging peer-to-peer (P2P) techniques in order to provide scalable, reliable microblogging services. The demo will show these aspects of Cuckoo and provide insights on the performance gain that decentralization and socioawareness can bring for microblogging systems.
Cuckoo_sigcomm10.pdf [257.5 kB]
Intrinsic Monitoring within an IPv6 Network: Mapping Node Specific information to Network Paths ,
Lei Shi , Alan Davy, David Muldowney, Steven Davy, Edzard Hoefig, Xiaoming Fu , International Conference on Network and Service Management (CNSM),
October 2010.
Improving Prediction Accuracy of Matrix Factorization Based Network Coordinate Systems ,
Yang Chen , Peng Sun, Xiaoming Fu , and Tianyin Xu , IEEE ICCCN 2010 Track on Multimedia and Peer-to-Peer Networking (MP2P),
IEEE, August 2010.
Read abstract
Network Coordinate (NC) systems provide a lightweight and useful way for scalable Internet distance prediction while serving as an important component in many Peer-to-Peer applications. Most of the existing NC systems utilize the Euclidean distance model, which is largely impaired by the persistent occurrence of Triangle Inequality Violation (TIV) in the Internet. Recently, matrix factorization (MF) based NC systems, which can completely remove the TIV constraint, provide an alternative approach towards better prediction accuracy. Phoenix, an NC system based on the MF model, well explores the advantage of the MF model and becomes the most accurate NC system so far. However, through experimental study, we find that the prediction accuracy of Phoenix for short links is significantly worse compared with the overall prediction accuracy. Based on the observation, we propose a new NC system, named Pancake, aiming at reducing the high prediction error for short links. By introducing a two-level architecture, Pancake achieves much higher prediction accuracy than other selected existing NC systems. Through extensive experiments, we demonstrate that Pancake reduces the 90th percentile relative error by up to 25.37% from Phoenix. Moreover, Pancake converges very fast and is robust to different dimension values. For further insights, we study the performance of Pancake using a dynamic data set in addition to the widely used aggregate data sets. With varying RTTs over time, Pancake outperforms other NC systems consistently.
Pancake_ICCCN10.pdf [269.1 kB]
NF-TCP: A Network Friendly TCP Variant for Background Delay-Insensitive Applications ,
Mayutan Arumaithurai , Xiaoming Fu , K. K. Ramakrishnan, Technical Report No. IFI-TB-2010-05, Institute of Computer Science, University of Goettingen, Goettingen, Germany,
ISSN 1611-1044, August 2010.
Read abstract
Delay-insensitive applications, such as P2P file sharing, generate
substantial amounts of traffic and compete with other applications on an equal
footing when using TCP. Further, to optimize throughput, such applications open
multiple connections. This results in an unfair and potentially poor service for
other applications having stringent performance objectives. In this paper, we
propose NF-TCP, a TCP variant for P2P and similar background delay-insensitive
applications. NF-TCP aims to be submissive to delay-sensitive applications under
congestion. It is designed to be network friendly based on a fluid flow
model for intermediate queues and uses explicit congestion notification
(ECN) for early detection of congestion. Moreover, it exploits the measure of the
available bandwidth to be able to aggressively utilize spare capacity.
We implemented NF-TCP on Linux and ns-2. Our evaluations of the NF-TCP Linux
implementation on ns-2 show that NF-TCP outperforms other network friendly approaches
(e.g., LEDBAT, TCP-LP and RAPID). NF-TCP achieves high utilization, fair bandwidth
allocation among NF-TCP flows and maintains a small average queue. Our evaluations
further demonstrate that with NF-TCP, the available bandwidth can be efficiently
utilized for supporting both delay-sensitive and insensitive traffic in a
wide range of scenarios.
PDF [5533.4 kB]
NF-TCP: Network Friendly TCP (Extended Abstract) ,
Mayutan Arumaithurai , Fabian Glaser , Xiaoming Fu , K. K. Ramakrishnan, 10th Würzburg Workshop on IP (EuroView2010): Visions of Future Generation Networks,
August 2010, Würzburg, Germany, August 2010.
ARA: A Routing and Forwarding Scheme for Coordinated Wide Area Mobility ,
Niklas Neumann , Xiaoming Fu , Gong Zhang, IEEE International Conference on Communications (ICC) 2010, Cape Town, South Africa,
IEEE, June 2010.
Read abstract
The wide availability of WiFi hot-spots provides mobile users with convenient access to the Internet. However, hot-spots cover only small areas and are often operated by a local authority. A user that roams between hot-spots effectively traverses different operators and access networks. Due to deployment complexity, IP mobility is usually not supported by the mobile node or the network. In this paper we propose an alternative approach which enables the access routers to provide seamless mobility support without any further changes to the network or the mobile node. Our approach provides a simple, flexible and deployable framework for mobility support and scales well with the number of involved access routers. Through simulation and analysis, we show that even with lower deployment requirements and less support from the network, our approach performs well in terms of routing stretch, handover latency and signaling overhead.
PDF [2673.6 kB]
D-MORE: Dynamic Mesh-based Overlay Peer-to-Peer Infrastructure ,
Jun Lei , Xiaoming Fu , and Dieter Hogrefe, Computer Communications: The International Journal for the Computer and Telecommunications Industry, Volume 33, Issue 10, Pages 1191-1201,
Elsevier, June 2010.
Read abstract
Traditionally, Internet Service Providers (ISPs) have to interconnect with content providers to provide network services to customers. The current business model that connectivity and bandwidth become commodities has motivated ISPs to distribute content and other application-specific services directly to their customers. Hence, it is desirable for ISPs to economize existing infrastructure to support a variety of applications and services.
We propose a dynamic mesh-based overlay peer-to-peer infrastructure (D-MORE) and illustrate its example usage case among other potentials. We describe several key techniques, namely capacity classification, locality-awareness and incentive mechanisms for construction of the tiered infrastructure. Through extensive simulations, we show D-MORE scales well with an increasing number of hosts, in terms of control overhead and link stress, for supporting media distribution services. We propose further improvements to enhance the D-MORE performance, which brings up to 35% network resource savings and up to 200% control overhead reduction in our simulations.
PDF [586.3 kB]
APEX: A Personalization Framework to Improve Quality of Experience for DVD-like Functions in P2P VoD Applications ,
Tianyin Xu , Baoliu Ye, Qinhui Wang, Wenzhong Li , Sanglu Lu, Xiaoming Fu , the 18th IEEE International Workshop on Quality of Service (IWQoS 2010), Beijing, China,
IEEE, June 2010.
Read abstract
The requirement for supporting DVD-like functions raises new challenges to the design of P2P VoD systems. The uncertainty of frequent user DVD-like interactivity makes it difficult to ensure user perceived Quality of Experience (QoE) for real-time streaming services over distributed self-organized P2P overlay networks. Most existing solutions are based on the unreasonable assumption that all the users in P2P VoD systems have the same preference. Few attention has been paid to personalization, which accommodates the differences between users. In this paper, we present a video model which characterizes the personalization information for users' contents and preferences. Based on this model, we develop APEX, a practical personalization framework for P2P VoD applications. APEX makes the personalization practical by using a hybrid architecture which leverages the offline pattern mining on the server side and online collaborative filtering on the peer side. Furthermore, APEX helps peers to personalize navigation, prefetching and membership management, aiming at improving QoE for DVD-like functions by reducing the response latency and optimizing content sharing. Both theoretical analysis and comprehensive simulations show that APEX outperforms most existing schemes in terms of accumulated hit ratio, response latency, and searching efficiency.
PDF [392.2 kB]
Towards Decentralized, Socio-Aware Online Microblogging Services and Data Measurement ,
Tianyin Xu , Yang Chen , Jin Zhao , and Xiaoming Fu , In Proc. ACM MobiSys 2010 HotPlanet Workshop,
ACM, June 2010.
Read abstract
Online microblogging services, as exemplified by Twitter and Yammer, have become immensely popular during the latest three years. Twitter, the most successful microblogging service, has attracted more than 41.7 million users as of July 2009 and is still growing fast. However,
current microblogging systems severely suffer from performance bottlenecks and central points of failure due to their centralized architecture. Thus, centralized microblogging systems may threaten the scalability, reliability, as well as availability of the offered services, not to mention the extremely high operational and maintenance cost.
However, it is not trivial to decentralize microblogging services in a peer-to-peer fashion. The challenges first derive from the heterogeneity of the inherent online social network (OSN) features. The non-reciprocation feature of microblogging services also increases the heterogeneity. Moreover, different from traditional approaches used in centralized server based systems, an efficient, robust and scalable approach for data collection and dissemination in such distributed heterogeneous environments is desirable.
In this paper, we present a decentralized, socio-aware microblogging system named Cuckoo. The design takes advantages of the inherent social relationships while leverages P2P techniques towards scalable, reliable microblogging services. Besides, Cuckoo provides a flexible interface for data collection while circumventing unnecessary traffic on the server. We discuss the benefits that our system may bring for both service providers and end users. We also discuss the technical aspects to be considered and report our work in progress.
Cuckoo_Hotplanet10.pdf [223.9 kB]
NF-TCP: Network Friendly TCP ,
Mayutan Arumaithurai , Xiaoming Fu , K. K. Ramakrishnan, 17th IEEE Workshop on Local and Metropolitan Area Networks (LANMAN 2010), Long Branch, New Jersey, USA,
IEEE, May 2010.
Read abstract
Delay-insensitive applications such as P2P file sharing, data center backups and software updates generate substantial amounts of traffic. This traffic, transported potentially over multiple TCP connections, competes with traffic from other possibly interactive applications. Today, with TCP, they compete on an equal level for each individual connection.
In this paper, we propose a new TCP variant for such delay-insensitive applications, which we call Network Friendly TCP
(NF-TCP). NF-TCP is responsive to available bandwidth, seeking to quickly and efficiently utilize the same in a congestion-free situation, while backing-off more aggressively than standard TCP on encountering competing traffic in a congested network. NF-TCP uses a novel combination of utilizing measurement of available bandwidth and ECN-based congestion avoidance techniques to ensure
that it is truly friendly to existing TCP connections.
We evaluate the performance of NF-TCP through ns-2 simulations and present the initial results enlightening the friendly nature of NF-TCP compared to standard TCP.
A Personalization Framework to Improve Quality of Experience for DVD-like Functions in P2P VoD Applications ,
Tianyin Xu , Baoliu Ye, Qinhui Wang, Wenzhong Li , Sanglu Lu, Xiaoming Fu , Technical Report No. IFI-TB-2010-03, Institute of Computer Science, University of Goettingen, Goettingen, Germany,
ISSN 1611-1044, May 2010.
Read abstract
The requirement for supporting DVD-like functions raises new challenges to the design of P2P VoD systems. The
uncertainty of frequent user DVD-like interactivity makes it difficult to ensure user perceived Quality of Experience
(QoE) for real-time streaming services over distributed self-organized P2P overlay networks. Most existing solutions
are based on the unreasonable assumption that all the users in P2P VoD systems have the same preference. Few
attention has been paid to personalization, which accommodates the differences between users. In this paper, we
present a video model which characterizes the personalization information for users’ contents and preferences. Based
on this model, we develop APEX, a practical personalization framework for P2P VoD applications. APEX makes
the personalization practical by using a hybrid architecture which leverages the offline pattern mining on the server
side and online collaborative filtering on the peer side. Furthermore, APEX helps peers to personalize navigation,
prefetching and membership management, aiming at improving QoE for DVD-like functions by reducing response
latency and optimizing content sharing. Both theoretical analysis and comprehensive simulations show that APEX
outperforms most existing schemes in terms of accumulated hit ratio, response latency, and searching efficiency.
PDF [506.9 kB]
LENS: Leveraging anti-social Networking against Spam ,
Sufian Hameed , Pan Hui, Xiaoming Fu , Technical Report No. IFI-TB-2010-02,
Institute of Computer Science, University of Göttingen, Germany, ISSN 1611-1044, April 2010.
Read abstract
Spam is still an open problem from the network operator's perspective. The common state-of-the-art strategy to place filters against spam is at the recipient's edge. Although this strategy largely solves the spam problem from the user's perspective--false positives/negatives may still exist--it cannot prevent spam from traversing the Internet. Consequently, with nowadays around $200$ billion spam/day, spam continues to consume large amounts of Internet bandwidth and provokes non-negligible financial loss to network operators. Therefore it becomes imperative to mitigate spam much earlier than at the recipient's edge. This goal has been recently accomplished only partially by placing filters at the edge of a social circle within a social network.
In this paper we introduce LENS, a novel spam protection system based on the anti-social networking paradigm, which further mitigates spam beyond social circles. The key idea of this paradigm in LENS is to let users select legitimate and authentic users, called Gatekeepers (GKs), from outside their social circle and within pre-defined social distances. Unless a GK vouches for the emails of potential senders from outside the social circle of a particular recipient, those e-mails are prevented from transmission. This way LENS drastically reduces the consumption of Internet bandwidth by spam to control messages only. To evaluate the scalability of LENS we use publicly available online social network (OSN) datasets and demonstrate that it is feasible to use GKs in the order of hundreds to provide reliable email delivery from millions of potential users. Using real email traces from large commercial and academic units, we demonstrate that LENS is very effective in accepting all inbound legitimate emails.
LENS [545.7 kB]
SybilConf: Computational Puzzles for Confining Sybil Attacks ,
Florian Tegeler , Xiaoming Fu , IEEE INFOCOM 2010 Student Workshop,
March 2010.
LEDBAT architecture framework consisting of pluggable components ,
Mayutan Arumaithurai , Xiaoming Fu , K. K. Ramakrishnan, IETF, draft-mayutan-ledbat-congestionarchitecture-00.txt,
March 2010.
Read abstract
The Low Extra Delay Background Transport (LEDBAT) working group is
considering protocols for an alternative congestion control protocol
that enables a delay-insensitive networking application to minimize
the extra queueing delay it causes to other applications because of
additional queueing at the bottleneck, when these connections
carrying traffic for such applications attempt to use the available
bandwidth.
This document proposes an architectural framework for LEDBAT
congestion control mechanisms, based on existing work on congestion
control protocols and the requirements of the LEDBAT working group.
The architectural framework consists of a LEDBAT-congestion control
(LEDBAT-CC) suite that provides flexibility in utilizing different
components for providing congestion control for transport connections
carrying delay-insensitive traffic. The LEDBAT-CC suite of protocols
is envisioned to support the multiple alternative mechanisms for
bandwidth estimation, congestion detection and indication and end-
system flow control to comprise a network friendly congestion
avoidance protocol.
This document is inspired by the need to standardize the various
components that constitute the network friendly congestion control
protocol to avoid having to individually standardize a multitude of
distinct and monolithic solutions.
Dimension Reduction of Network Bottleneck Bandwidth Data Space ,
Peng Sun, Yang Chen , Yibo Zhu, Xiaoming Fu , Beixing Deng, Xing Li, Proc. of the 29th IEEE Conference on Computer Communications (INFOCOM'10) Student Workshop,
March 2010.
Read abstract
The network proximity metrics, such as bottleneck
bandwidth and round-trip time, are very useful in different
network applications. The round-trip-time prediction has been studied extensively. However, the prediction of bottleneck bandwidth has received much less attention. Therefore, we attempt to design a new bottleneck bandwidth prediction system by matrix factorization. As a first step, we focus on the dimension reduction of network bottleneck bandwidth data space in this paper. Evaluation is carried out based on real-world bottleneck bandwidth datasets, which are collected in the past three months. The results show that a 250D data space can be compressed to 10D and the average median-relative-error is only 8.65%. Although preliminary, our work provides some insights into the
design direction towards matrix factorization based distributed system to predict the bottleneck bandwidth.
BW_INFOCOM10SW.pdf [77.7 kB]
A Cross-Layer Approach for Improving TCP Performance in Mobile Environments ,
Deguang Le , Xiaoming Fu , and Dieter Hogrefe, Wireless Personal Communications, 52(3): 669-692,
Springer Verlag, ISSN 0929-6212, February 2010.
Read abstract
Network-layer mobility protocols have been developed to keep continuous connectivity for mobile hosts while transparent to the higher layers. However, Due to its distinct characteristics of different from traditional TCP/IP environment, mobility poses substantial impacts on TCP performance in mobile environments.
This paper proposes a new cross-layer approach, by introducing a mobility detection element in the network layer which interacts with the transport layer to optimize TCP operations. As changes are only made to the endpoints, this approach preserves the end-to-end semantics of TCP. Different from most exiting works, which utilize either transport or network layer alone without much cross-layer cooperation, our approach allows the use of mobility information in TCP. We analytically compare this approach against existing approaches and show that our approach outperforms prior approaches in terms of effective data resumption time. Through performance simulations, our approach demonstrates that it can effectively improve TCP performance in Mobile IPv6-based mobile environments.
PDF [758.2 kB]
Annual Report of Computer Networks Group at the University of Göttingen (2009) ,
Xiaoming Fu (ed.),
January 2010.
Read abstract
This document reports the major activities performed and results achieved by the Computer Networks Group at the University of Göttingen throughout the calendar year 2009.
PDF [105.5 kB]
2009
Pre-Congestion Notification based Flow Management in MPLS-based DiffServ Networks ,
Mayutan Arumaithurai , Ruediger Geib, Rene Rex , and Xiaoming Fu , The 28th IEEE International Performance Computing and Communications Conference (IPCCC 2009), Phoenix, AZ, USA,
IEEE, December 2009.
Read abstract
Recent trends in applications require per-flow treatment instead of the packet-based treatment inherent in the present day Internet. This requirement is analogous to trying to make a
phone call in a standard PSTN line or cellular network and either receiving a busy tone or call accept. This paper explores the potential of providing flow-based
admission control and termination functionality to an MPLS-based DiffServ network by applying Pre-congestion notification (PCN). MPLS-based DiffServ
networks are state of the art technologies used in many carrier backbones to provide Quality of Service for different traffic classes. The standard PCN scheme as developed by the IETF
requires three code-points, one of which finds application in differentiating between PCN and non-PCN traffic, while the other two are used for PCN congestion marking to indicate admission control
and flow termination, respectively. An MPLS-based DiffServ network on the other hand is limited as only two code-points are available. To overcome this limitation, we propose various schemes,
namely: 1) one-codepoint signaling for admission control (AC), 2) one-codepoint
signaling for flow termination (FT), and 3) one-codepoint signaling for 3
states (no congestion, AC and FT). The three schemes use the
available code-point for different purposes and require different behaviors from
the involved PCN nodes. We evaluate them using simulations.
The results suggest that the third approach is able to effectively signal the three states without incurring much
inaccuracy. The success rate with respect to the Quality of Experience (QoE) of existing flows, error recovery and throughput is not compromised either. To the
best of our knowledge, this is the first known attempt to study the use of PCN in an MPLS-based DiffServ network.
ipccc1569246747.pdf [183.2 kB]
End-to-End Versus Hop-by-Hop Soft State Refresh for Multi-hop Signaling Systems ,
Jianhua He, Xiaoming Fu and Zuoying Tang, 17th IEEE International Conference on Network Protocols (ICNP 2009), Princeton, New Jersey, USA,
IEEE, October 2009.
Read abstract
To ensure state synchronization of signalling operations, many signaling protocol designs choose to establish “soft” state that expires if it is not refreshed. The approaches of refreshing state in multi-hop signaling system can be classified as either end-to-end (E2E) or hop-by-hop (HbH). Although both state refresh approaches have been widely used in practical signaling protocols, the design tradeoffs between state synchronization and signaling cost have not yet been fully investigated. In this paper, we investigate this issue from the perspectives of state refresh and state removal. We propose simple but effective Markov chain models for both approaches and obtain closed-form solutions which depict the state refresh performance in terms of state
consistency and refresh message rate, as well as the state removal performance in terms of state removal delay. Simulations verify the analytical models. It is observed that the HbH approach yields much better state synchronization at the cost of higher signaling cost than the E2E approach. While the state refresh performance can be improved by increasing the values of state refresh and timeout timers, the state removal delay increases largely for both E2E and HbH approaches. The analysis here shed lights on the design of signaling protocols and the configuration of the timers to adapt to changing network conditions.
PDF [176.2 kB]
Fast Rerouting for IP Multicast in Managed IPTV Networks ,
Ralf Lübben , Guangzhi Li, Dongmei Wang, Robert Doverspike, and Xiaoming Fu , IEEE International Workshop on Quality of Service (IWQoS 2009), Charleston, SC, USA,
July 2009.
Read abstract
Recent deployment of IP based multimedia distribution, especially broadcast TV distribution has increased the importance of simple and fast restoration during IP network failures for service providers. In this paper, we propose and evaluate a simple but efficient method for fast rerouting of IP multicast traffic during link failures in managed IPTV networks. More specifically, we devise an algorithm for tuning IP link weights so that the multicast routing path and the unicast routing path between any two routers are failure disjoint, allowing us to use unicast IP encapsulation for undelivered multicast packets during link failures. We demonstrate that, our method can be realized with minor modification to the current multicast routing protocol (PIM-SM). We run our prototype implementation in Emulab which shows our method yields to good performance.
PDF [166.6 kB]
Method for carrying out a QoS-oriented handoff between a first and a second IP-based, especially mobile IPV6-based, communication path, between a mobile node (MN) and a correspondent node (CN) ,
Changpeng Fan, Andreas Festag, Xiaoming Fu , Cornelia Kapper, Holger Karl, Mirko Schramm, and Günter Schäfer (inventors), granted patent, No. EP20010953886,
July 2009.
Read abstract
The invention relates to a method for carrying out a QoS-oriented handoff between a first and a second IP-based, especially mobile IPv6-based, communication path, between a mobile node (MN) and a correspondent node (CN), the second communication path being part of a number of communication paths which can be accessed by the mobile node, with no, one, or a plurality of intermediate instances. The inventive method comprises at least the following steps: (a) a communication path is selected from the communication paths which can be accessed by the mobile node, as a second communication path; (b) a message (BU) is generated by the mobile node, said message containing at least one IP address which is associated with the mobile node on the basis of the selected communication path, and containing minimum quality of service requirements (QoS) in terms of the selected communication path; (c) the ability to meet at least the minimum quality of service requirements is controlled and optionally ensured by the individual intermediate instances through which the message passes successively, on the selected communication path and/or through the correspondent node. The message contains the minimum quality of service requirements for a communication from the mobile node to the correspondent node and/or vice versa. A handoff is automatically carried out between the first communication path and the second selected communication path, when at least the minimum quality of service requirements are met or the message is stopped. A notice is generated in an intermediate instance and/or in the correspondent node and is sent to the mobile node if the ability to meet the minimum quality of service requirements is not ensured.
Loop-Free Forwarding Table Updates with Minimal Link Overflow ,
Lei Shi , Jing Fu, and Xiaoming Fu , Proceedings of IEEE International Conference on Communications (ICC 2009), Dresden, Germany,
June 2009.
Read abstract
The forwarding paths in an IP network may change due to a link failure, network equipment maintenance or reconfiguration of link weights, then the forwarding tables in the routers need to be updated. These updates may cause transient loops and transient link overflow, if they are not performed in an appropriate order. While existing work has been done to achieve loop-free updates, transient link overflow is still a problem during the update process. In this paper, we present a method that compares the initial and final forwarding paths, and obtains the updatable nodes that do not cause any transient loop or transient link overflow. However, there is not always such kind of nodes so that the forwarding tables may not converge to the final one without causing link overflow. Therefore, we propose an algorithm to update the forwarding tables that will refrain the link overflows to a minimal level. The performance study on a real topology with two setups confirms that our approach achieves smaller link overflow than using a previously proposed approach.
PDF [219.0 kB]
TORI: User Provided Future Networking Testbeds ,
Martin Stiemerling , Marcus Brunner, Sebastian Kiesel, and Xiaoming Fu , IEEE International Workshop on the Network of the Future, in conjunction with IEEE ICC 2009, Dresden, Germany,
IEEE, June 2009.
Read abstract
The usage of testbeds is considered a key tool for exploring the development of new protocols and network architectures in the area of network research. Testbeds, together with simulations, are the basic tool set of network researchers to drive research, but often it is impossible to get feedback from real deployments and their respective data traffic. Today’s major testbed facilities, e.g., VINI and PlanetLab, aim at emulating the behavior of large-scale networks, but they are still several orders of magnitude smaller than the deployed operational network infrastructure. We argue that it is time to extend network research beyond theoretical and testbed approaches towards a dynamic, peer-to-peer based testbed environment, similar to the approach taken by seti@home and BOINC. We aim at expanding the total number of participating nodes in an experiment and at experimenting on existing operational infrastructure with its entirely uncontrollable environment. Our vision presented in this paper, the Testbed on Real Infrastructure (TORI), includes regular end hosts (peers) in an experiment by deploying and executing the experimental software on these peers and to form an overlay network upon them. The main difference of our TORI approach compared to others is installing new technologies and testing them with the operational infrastructure.
tori-final.pdf [141.2 kB]
Interest-based Peer-to-Peer Group Management ,
Jun Lei , Xiaoming Fu , Second IEEE/ACM International Workshop on Future Multimedia Networking (FMN 2009), Coimbra, Portugal,
Springer LNCS, June 2009.
Read abstract
Peer-to-Peer systems become popular applications but suffer from insufficient resource availability which is caused by free-riders and inefficient lookup algorithms. To address the first cause, a number of recent works have focused on providing appropriate incentive mechanisms to encourage participants to contribute their resources to the P2P systems. To improve the lookup efficiency, locality-aware peer management has been introduced into the research community. However, existing proposals attempt to optimize the service performance during the data transmission period mostly after performing the neighboring lookup, which cannot address the fundamental concern of reducing lookup traffic. Besides, existing implementations select available contributors among random neighbors suggested by a specific server. Therefore, this paper proposes interest-based peer-to-peer management (IPM) protocol to facilitate the peering lookup. Our design philosophy differs from existing work that IPM is a client-only approach and can be represented as either an alternative or a complementary to the current proposals. With additional locality-awareness considerations, IPM can reduce the lookup overhead while optimizing the P2P traffic performance. The simulation results essentially state that IPM can largely improve the efficiency and reliability of P2P media distribution systems, for instance, reduces control overhead by 50% on average and reduces average packet loss rate up to 34.7%.
PDF [234.0 kB]
I-PMIP: An Inter-Domain Mobility Extension for Proxy Mobile IP ,
Niklas Neumann , Jun Lei , Xiaoming Fu , Gong Zhang, in the Proceedings of 5th International Wireless Communications and Mobile Computing Conference (IWCMC 2009), Leipzip, Germany,
ACM Digital Library, June 2009.
Read abstract
Proxy Mobile IP (PMIP) provides a solution for network-based localized mobility management which in contrast to host-based mobility solutions, like Mobile IP (MIP), does not require changes to the end-hosts and avoids tunneling overhead between the mobile node and it's network access point. Within a PMIP-enabled mobility domain, the mobile node is able to maintain the same IP address when it moves. However, if the mobile node leaves this domain the mobility support breaks. This paper proposes an extension to PMIP, called I-PMIP which allows to interconnect multiple PMIP-enabled mobility domains to provide continuous mobility support for a mobile user. I-PMIP is based on an architecture that can provide a mobile node with an anchor point that is placed very close towards the mobile. Numerical analysis show that the approach is comparable to other approaches that provide inter-domain mobility.
PDF [231.4 kB]
XOR Rescue: Exploiting Network Coding in Lossy Wireless networks ,
Fang-Chun Kuo , Kun Tan, Xiang-Yang Li, Jiansong Zhang and Xiaoming Fu , 6th IEEE Communications Society Conference on Sensor, Mesh and Ad Hoc Communications and Networks (SECON 2009), Rome, Italy,
June 2009.
Read abstract
It is well-known that wireless links are error-prone and require retransmissions for recovering frames from errors and losses. Network coding (NC) has been proposed for more efficient MAC-layer retransmissions in WLANs. However, existing schemes employed the reception report mechanism, which is both inefficient and expensive. Furthermore, they considered neither fairness nor the effects of time-varying heterogeneous wireless networks. These issues are critical for achieving full benefit of network coding. Without addressing them, these schemes may even impair system performance. In this paper, a novel MAC-layer retransmission scheme, namely XOR Rescue(XORR) is proposed. It estimates the reception status without extra overheads and devises a new coding metric, which accommodates the effects of the frames size and the channel condition. Finally, XORR employs NC-aware fair opportunistic scheduling, which is theoretically proven to be fair, i.e. not only the service time is evenly allocated, but also it always improves the expected goodput for every wireless station. It is further verified by theoretic analyses, extensive simulations and testbed experiments. Our results show that XORR outperforms the non-coding fair opportunistic scheduling and 802.11 by 25% and 40%, respectively.
PDF [307.7 kB]
Proceedings of the 1st ACM International Workshop on Hot Topics of Planet-scale Mobility Measurement (HotPlanet'09) ,
Xiaoming Fu , Pan Hui (editors), Krakow, Poland,
ACM Press, ISBN 978-1-60558-689-2, June 2009.
Routing and Scheduling for WiMAX Mesh Networks ,
Jianhua He, Xiaoming Fu , Jie Xiang, Yan Zhang and Zuoyin Tang, in: Y. Zhang (ed.), WiMAX Network Planning and Optimization, pages 221-243,
Auerbach Publications, Taylor&Francis Group, USA, ISBN 978-1-4200-6662-3, April 2009.
Read abstract
The book chapter presents the recent developments and a new approach for the scheduling and (both distributed and centralized) routing issues in WiMAX mesh networks.
PDF [408.7 kB]
End-to-End versus Hop-by-Hop State Refresh in Soft State Signaling Protocols ,
Jianhua He, Xiaoming Fu , Zuoyin Tang, and Hisiao-Hwa Chen, IEEE Communications Letters, 13(4): 268-270,
IEEE, April 2009.
Read abstract
The concept of soft state (i.e., the state that will expire unless been refreshed) has been widely used in the design of network signaling protocols. The approaches of refreshing state in multi-hop networks can be classified to end-to-end (E2E) and hop-by-hop (HbH) refreshes. In this article we propose an effective Markov chain based analytical model for both E2E and HbH refresh approaches. Simulations verify the analytical models, which can be used to study the impacts of link characteristics on the performance (e.g., state synchronization and message overhead), as a guide on configuration and optimization of soft state signaling protocols.
PDF [168.3 kB]
A Unified Security Backplane for Trust and Reputation Systems in Decentralized Networks ,
Florian Tegeler , Jun Lei , and Xiaoming Fu , IEEE INFOCOM 2009 Student Workshop,
April 2009.
Read abstract
Trust and Reputation (TR) systems are a recently proposed means to address free-rider issues in decentralized networks such as P2P, DTNs, and wireless mesh networks. Basically, TR systems identify malicious node behaviors by observation and direct interaction experience. However, these systems often lack a security framework to prevent a variety of attacks, such as identity spoofing or capturing and false reports on nodes behavior. We present a security backplane preventing such attacks by providing authentication, non-repudiation and other security services without predetermining the exact TR algorithm on node interaction selection and the metrics on the evaluation of nodes. Utilizing this security framework, multiple proposed algorithms such as Scrubber, Eigentrust, CONFIDENT or pricing systems can be implemented with increased and flexible security properties.
PDF [79.8 kB]
An encoding method to signal 3 states with a single PCN bit ,
Mayutan Arumaithurai , Ruediger Geib, Rene Rex , and Xiaoming Fu , IEEE INFOCOM 2009 Student Workshop,
April 2009.
Read abstract
Pre-Congestion Notification (PCN) is currently being developed by the IETF to provide admission control in DiffServ networks for non-elastic flows. Various marking schemes are being proposed as part of this effort. We evaluate a new marking behaviour that could be used to signal three different states by means of a single bit. This scheme is of interest to networks operating with MPLS as the Label
QoS coding space is limited to three EXP bits. Similar to the ECN standard for MPLS, this research assumes that just two codepoints will be available to indicate different congestion states. Of these two codepoints, one would be used to differentiate between PCN and non-PCN traffic and the other would be used for PCN marking. This paper outlines the approach and presents the advantages and limitations of the proposal using the evaluations performed. A comparison of the method with a two bit PCN marking approach is also studied.
pcn-markingPoster.pdf [106.3 kB]
Overhead and Performance Study of the General Internet Signaling Transport (GIST) Protocol ,
Xiaoming Fu , Henning Schulzrinne, Hannes Tschofenig , Christian Dickmann , and Dieter Hogrefe, ACM/IEEE Transactions on Networking, 17(1): 158-171,
February 2009.
Read abstract
The General Internet Signaling Transport (GIST) protocol is currently being developed as the base protocol component in the IETF Next Steps In Signaling (NSIS) protocol stack to support a variety of signaling applications. We present our study on the protocol overhead and performance aspects of GIST. We quantify network-layer protocol overhead and observe the effects of enhanced modularity and security in GIST. We developed a first open source GIST implementation at the University of Goettingen, and study its performance in a Linux testbed. A GIST node serving 45,000 signaling sessions is found to consume average only 1.1 ms for processing a signaling message and 2.4 KB of memory for managing a session. Individual routines in the GIST code are instrumented to obtain a detailed profile of their contributions to the overall system processing. Important factors in determining performance, such as the number of sessions, state management, refresh frequency, timer management and signaling message size are further discussed. We investigate several mechanisms to improve GIST performance so that it is comparable to an RSVP implementation.
PDF [210.7 kB]
Biannual Report of Computer Networks Group at the University of Göttingen (2007-08) ,
Xiaoming Fu (ed.),
February 2009.
Read abstract
A (bi)annual report series for the Computer Networks Group at the University of Göttingen intends to review the most relevant and important research results and other achievements and activities performed by the research group. The present report summarizes 2007-2008, the first period that the group got launched.
PDF [137.9 kB]
2008
Diameter WebAuth: An AAA-based Identity Management Framework for Web Applications ,
Niklas Neumann , and Xiaoming Fu , The 51th Annual IEEE Global Telecommunications Conference (GLOBECOM 2008), Computer and Communications Network Security Symposium, New Orleans, LA, USA,
IEEE, December 2008.
Read abstract
With an increasing number of personalized information and services offered on the Internet, especially the World WideWeb, effective identity management solutions are demanded by application providers. Instead of a web-based stand-alone approach, we extend existing network-based AAA mechanisms to be usable for identity management by web applications. Our proposal, Diameter WebAuth, allows to seamlessly integrate web-based services into a Diameter infrastructure for authentication, authorization, credit-control and identity management purposes. Diameter WebAuth offers comparable features to web-based identity management solutions, benefits from the maturity and wide deployment of the Diameter protocol, and takes advantage of existing AAA setups.
PDF [253.4 kB]
Performance Study of the NSIS QoS-NSLP Protocol ,
Mayutan Arumaithurai , Xiaoming Fu , Bernd Schloer , and Hannes Tschofenig , The 51th Annual IEEE Global Telecommunications Conference (GLOBECOM 2008), Next Generation Networks, Protocols, and Services Symposium, New Orleans, LA, USA,
IEEE, December 2008.
Read abstract
This paper presents an evaluation of the Quality of Service Signalling Layer Protocol (QoS-NSLP) of the NSIS (Next Steps In Signalling) protocol suite. The QoS-NSLP in combination with the NSIS Transport Layer Protocol (NTLP) is proposed by the Internet Engineering Task Force (IETF) as an alternative to the Resource reSerVation Protocol (RSVP). We describe our implementations of the software architecture, both on a network simulator and on a Linux implementation. Both implementations are used in a complimentary manner to illustrate the performance of the QoS-NSLP protocol. The results show the performance of QoS-NSLP with respect to resource consumption, packet processing time, session set up time, refresh interval and protocol overhead. Furthermore, we analyse the protocol performance during route change scenarios.
PDF [328.7 kB]
Forschungsrichtungen der Internettechnologie - Kein Tempolimit für die Datenautobahn ,
Xiaoming Fu , Dieter Hogrefe, and Henning Schulzrinne, Georg Augusta, special issue on Mathematics and Computer Science Year 2008, Vol. 6, 112-118, pages ,
ISSN 0016-8157, December 2008.
Read abstract
Selected research fields on Internet technologies are presented and open issues and ongoing topics, as a result of the discussions in the 1st Columbia-Göttingen Workshop on Internet Research.
PDF [301.9 kB]
Decoupling Congestion Control Using Traffic Aggregates and Middleboxes ,
Niklas Neumann , Ralf Lübben , Mayutan Arumaithurai , and Xiaoming Fu , IEEE International Conference on Network Protocols (ICNP 2008), poster session, Orlando, FL, USA,
October 2008.
Read abstract
A rise in numbers of large bandwidth-delay product links and an increasing heterogeneity of IP networks bring new challenges for the existing congestion control mechanisms. Congestion control mechanisms are traditionally end-to-end oriented. This makes them slow to react on high delay links and inaccurate if the flow traverses heterogeneous network segments that have different jitter, or packet loss rates. Furthermore, the slow start which TCP uses as part of its congestion control is slow to react on high bandwidth-delay product links and makes it hard for short-lived flows to develop fully.
We propose to decouple the end-to-end congestion control mechanisms by introducing middleboxes to create dedicated congestion control segments within the network. Within those segments congestion control mechanisms can be deployed that are specifically adapted to the particular properties of this segment.
PDF [198.7 kB]
Implications and Control of Middleboxes in the Internet ,
Xiaoming Fu , Martin Stiemerling , and Henning Schulzrinne, IEEE Network, Special Issue on Implications and Control of Middleboxes in the Internet,
September 2008.
Read abstract
Middleboxes in the Internet have been explored, sometimes quite controversially, in operations, standardization, and the research community for more than 10 years. The main concern, on one hand, has been their contradicting nature to the Internet's end-to-end principle. On the other hand, middleboxes were introduced in the Internet for various reasons. In this special issue we are pleased to introduce a series of state-of-the-art articles on this specific area. These articles cover the subject from a variety of perspectives, offering the readers an understanding of the issues and implications of various middleboxes in the Internet, including their control mechanisms.
PDF [140.6 kB]
Evaluating the benefits of introducing PMIPv6 for localized mobility management ,
Jun Lei , and Xiaoming Fu , in the Proceedings of International Wireless Communications and Mobile Computing Conference 2008 (IWCMC 2008), Crete, Greece, Pages 74-80,
IEEE, August 2008.
Read abstract
Abstract—Since recent years, it has been recognized that using global mobility protocol for managing localized mobility causes a number of problems, such as a long registration delay. To overcome these problems, Proxy Mobile IPv6 is proposed, which can avoid tunneling overhead over the air and support for hosts without an involvement in the mobility management. In this paper, we first discuss the recent localized mobility proposals and explore three major benefits that PMIPv6 can bring. In particular, we evaluate two aspects of the handover performance through a simple mathematical model for Fast Handovers for MIPv6, Hierarchical MIPv6, Fast handovers for HMIPv6 and PMIPv6. These analytical studies show that PMIPv6 may cause high handover latency if the local mobility anchor is located far from the current mobility access gateway. Therefore, some enhancements for PMIPv6 are suggested to further reduce the handover latency. The analysis ascertains that F-PMIPv6 is a promising mobility scheme to efficiently manage the localized mobility.
PDF [206.1 kB]
D-MORE: Dynamic Mesh-based Overlay Peer-to-Peer Infrastructure ,
Jun Lei , Xiaoming Fu , and Dieter Hogrefe, Technical Report No. IFI-TB-2008-02, Institute of Computer Science, University of Goettingen, Goettingen, Germany,
ISSN 1611-1044, August 2008.
Read abstract
Traditionally, Internet Service Providers (ISPs) have
to interconnect with content providers to provide network services
to customers. Current business model that connectivity
and bandwidth become commodities has motivated ISPs to
distribute content and other application-specific services to their
customers using their own infrastructure. It is desirable for
ISPs to economize existing infrastructure to support a variety
of applications and services.
We propose a dynamic mesh-based overlay peer-to-peer infrastructure
and illustrate its two examples usage cases among other
potentials. We describe several key techniques, namely capacity
classification, locality-awareness and incentive mechanisms for
construction of the tiered infrastructure. Through extensive
simulations, we show D-MORE scales well with an increasing
number of hosts, in terms of control overhead, link stress and
data path length, for supporting media distribution services.
We propose further improvements to enhance the D-MORE
performance, which brings up to 35% network resource savings
and up to 200% control overhead reduction in our simulations.
PDF [539.6 kB]
Network coding-aware fair opportunisic scheduling in wireless networks ,
Fang-Chun Kuo , Kun Tan, Xiang-Yang Li, Jiansong Zhang, and Xiaoming Fu , Technical Report No. IFI-TB-2008-03, Institute of Computer Science, University of Goettingen, Goettingen, Germany,
ISSN 1611-1044, August 2008.
Read abstract
Users increasingly depend on WLAN for business and entertainment. It is well-recognized that wireless links are prone to errors. Previous work, ER, proposed to use network coding (NC) for providing more efficient MAC-layer
retransmission scheme in WLAN. However, it uses inefficient and costly reception report scheme and does not consider the effect of heterogeneous and time-varying wireless conditions and fairness. These issues are critical for getting full benefits of network coding. We show that, without addressing them, NC may even cause negative effect on the system. In this paper, we present a novel MAC-layer retransmission scheme, namely XORR, which uses reception estimation without extra overhead and adopts NC-aware opportunistic scheduling with maintaining temporal fairness in WLAN. We prove our NC-aware scheduling algorithm is fair and it will always improve the expected goodput for each wireless clients. We further verify XORR with extensive simulation as well as experiment studies and find that our scheme outperforms traditional opportunistic scheduling (without NC) and 802.11 about 25% and 40%, respectively.
PDF [476.7 kB]
A Network Virtualisation Concept Based on Ambient Networks SATO System ,
Martin Stiemerling , Xiaoming Fu , and Marcus Brunner, 1. GI/ITG Fachgespraech Virtualisierung, Paderborn, Germany, pages 33 - 36,
February 2008.
Read abstract
Network virtualization can be one way of fixing the shortcomings of todays Internet but also open the venue for new, unforeseen applications. In this extended abstract, we present a novel approach for network virtualisation based on the Service-Aware Transport Overlay (SATO) concept of Ambient Networks. SATOs introduce on-demand overlay creation and new interfaces to ease applications to use overlays.
PDF [295.8 kB]
Reports of Advanced Topics in Computer Networking and Mobile Communications (Summer 2004 - Winter 2006/2007) ,
Xiaoming Fu , and Dieter Hogrefe (editors), Technical Report No. IFI-TB-2008-01, Institute of Computer Science, University of Goettingen, Goettingen, Germany,
ISSN 1611-1044, February 2008.
Read abstract
This technical report includes the final reports written by students for seminars from summer semester 2004 to winter semester 2006/07 on advanced topics in computer networking/Internet research and mobile communications, including mobile, ad hoc, sensor networks and location-based systems; wireless networks, security and performance optimization issues; mobile applications and mobile devices; overlay, peer-to-peer and application layer multicast, as well as delay tolerant networks.
PDF [8327.7 kB]
Probe-aided MulTCP: An Aggregate Congestion Control Mechanism ,
Fang-Chun Kuo , and Xiaoming Fu , ACM SIGCOMM Computer Communication Review, Vol. 38, No. 1, Pages 19-28,
ACM, ISSN 0146-4833, 2008.
Read abstract
An aggregate congestion control mechanism, namely Probe-Aided MulTCP (PA-MulTCP), is proposed in this paper. It is based on MulTCP, a proposal for enabling an aggregate to emulate the behavior of multiple concurrent TCP connections. The objective of PA-MulTCP is to ensure the fair sharing of the bottleneck bandwidth between the aggregate and other TCP or TCP-friendly flows while keeping lightweightness and responsiveness. Unlike MulTCP, there are two congestion window loops in PA-MulTCP, namely the probe window loop and the adjusting window loop. The probe window loop constantly probes the congestion situation and the adjusting window loop dynamically adjusts the congestion window size for the arriving and leaving flows within the aggregate. Our simulations demonstrate that PA-MulTCP is more stable and fairer than MulTCP over a wide range of the weight N in steady conditions as well as in varying congestion conditions. PA-MulTCP is responsive to flow-arriving/leaving and thus reduces the latency of short-lived transfers. Furthermore, PA-MulTCP is lightweight, since it enjoys above advantages at the cost of only an extra probe window loop, which has a marginal influence on the implementation complexity. Finally, the design of PA-MulTCP decouples the congestion management from the other functionalities in the aggregate flow management. As a result, PA-MulTCP could be potentially applied to a wider range of scenarios, e.g. wireless TCP proxies, edge-to-edge overlays, QoS provisioning and mass data transport.
PDF [758.8 kB]
2007
Optimized FMIPv6 Using IEEE802.21 MIH Services in Vehicular Networks ,
Qazi Mussabbir, Wenbing Yao, Zeyun Niu, and Xiaoming Fu , IEEE Transactions on Vehicular Technology, Special Issue on Vehicular Communications Networks, Volume 56, Issue 6, Pages 3397 - 3407,
IEEE, ISSN 0018-9545, November 2007.
Read abstract
In this paper, we optimize the handover procedure in Fast Handover for Mobile IPv6 (FMIPv6) protocol by using IEEE 802.21 Media Independent Handover (MIH) services. FMIPv6 is used to enhance the performance of handovers in Mobile IPv6 (MIPv6) and its basic extension for Network Mobility (NEMO), the fundamental mobility management protocols used in vehicular networks. With the aid of the lower three layers information of the mobile node/router (MN/MR) and the neighboring access networks, we tackle the radio access discovery and candidate Access Router (AR) discovery issues of FMIPv6. We introduce an Information Element Container to store static and dynamic Layer 2 (L2) and Layer 3 (L3) information of neighboring access networks, and propose to use a special cache maintained by the MN/MR to reduce the anticipation time in FMIPv6, thus increasing the probability of the predictive mode of the FMIPv6 operation. Furthermore, we propose a cross-layer mechanism for making intelligent handover decisions in FMIPv6. Lower layer information of the available links obtained by MIH services as well as the higher layer information such as quality of service parameter requirements of the applications are used by a Policy Engine (PE) to make intelligent handover decision. We will show through analysis and simulations of the signaling procedure that the overall expected handover (both L2 and L3) latency in FMIPv6 can be significantly reduced in the proposed mechanism.
PDF [214.6 kB]
A New Decentralized Mobility Management Service Architecture for IPv6-based Networks ,
Deguang Le , Jun Lei , and Xiaoming Fu , in Proceedings of the 3rd ACM International Workshop on Wireless Multimedia Networking and Performance Modeling (WMuNeP'07), in conjunction with the 10th ACM/IEEE International Symposium on Modeling, Analysis and Simulation of Wireless and Mobile Systems (MSWiM'07), Chania, Crete Island, Greece,
ACM Press, New York, NY, USA, October 2007.
Read abstract
In Mobile IPv6, the home network - through a designated home agent - is responsible for distributing all traffic from/to the mobile node in the default bidirectional tunneling mode, when the mobile node is connected to a foreign network. This approach not only lacks sufficient scalability and efficiency of delivery, but also poses a heavy burden on the home network and the global Internet. In this paper we propose a new decentralized mobility management service (DMMS) architecture to address this issue. The idea is to employ a local mobility agent in each access network, which handles node mobility based on local movement information, so that the ongoing communication can be maintained efficiently and scalable without relying on centralized traffic distributing entities.
PDF [297.9 kB]
An NSIS-based Approach for Firewall Traversal in Mobile IPv6 Networks ,
Niklas Steinleitner , Xiaoming Fu , Dieter Hogrefe, Thomas Schreck, and Hannes Tschofenig , Third Annual International Wireless Internet Conference (WICON 2007), Austin, Texas, USA,
ACM Press, October 2007.
Read abstract
Firewalls have been successfully deployed in todays network infrastructure in various environments and will also be used in IPv6 networks. However, most of the current firewalls do not support Mobile IPv6, the best known standardized solution for mobility support in IPv6. As a result, Mobile IPv6 traffic will be most likely dropped when used without an appropriate firewall traversal solution.
This paper describes the problems and impacts of having firewalls in Mobile IPv6 environments and presents a firewall traversal solution based on the IETFs Next Steps In Signaling framework to address these issues. Compared with other candidates such as STUN, TURN, ICE, ALG, MIDCOM
and COPS, this approach does not rely on specific firewall placements and can be applied in various operational modes without additional introducing entities. In this paper we also explore security aspects since they are typically difficult to handle.
PDF [372.2 kB]
An Experimental Analysis of Joost Peer-to-Peer VoD Service ,
Jun Lei , Lei Shi , and Xiaoming Fu , Technical Report No. IFI-TB-2007-03, Institute of Computer Science, University of Goettingen, Goettingen, Germany,
ISSN 1611-1044, October 2007.
Read abstract
Most of the current Video-on-Demand (VoD) systems rely on content distribution networks or some local streaming proxies. While these traditional systems offer a means for media delivery and streaming, they also pose a significant performance challenge in terms of scalability and service delay as the number of clients increases. To solve this issue, peer-to-peer (P2P) technologies have been applied to support the VoD systems. Joost is one of such systems for distributing TV shows or other forms of video over the Internet. However, like Skype in its early stage, the mechanisms behind Joost are still unrevealed.
The main purpose of this paper is therefore to study the underlying Joost architecture and its key components, and analyze its media streaming behaviors and peer management mechanisms through close investigations on Joost network traffic. With three envisioned typical scenarios we have further studied the Joost performance in terms of locality awareness, bandwidth capacity and VoD functionalities. Based on extensive experiments, we infer that Joost is a server-assisted peer-to-peer VoD system. It mainly relies on a set of delicate infrastructure nodes (e.g. content servers) for video distribution. To our best knowledge, this paper is the first analytical and performance study on commercial P2P VoD services.
PDF [436.4 kB]
An overview of digital TV standards in China ,
Roland A. Burger, Giovanni Iacovoni, Cliff Reader, Xiaoming Fu , Xiaodong Yang , and Wang Hui, Proceedings of ChinaCom 2007, Shanghai, China, Special Session on Digital Broadcasting and Mobile Convergence,
IEEE, August 2007.
Read abstract
This paper presents an updated overview of the different proposed standards on the market for mobile TV in China as of June 2007 and analyzes the different trade-offs, strengths and weaknesses. Furthermore the used video codecs are compared with special emphasis on the usage in mobile TV in China.
PDF [4370.5 kB]
Comparative Studies on Authentication and Key Exchange Methods for 802.11 Wireless LAN ,
Jun Lei , Xiaoming Fu , Dieter Hogrefe, and Jianrong Tan, Computers & Security, Volume 26, Issue 5, Pages 401-409,
Elsevier, ISSN 0167-4048, August 2007.
Read abstract
IEEE 802.11 wireless LAN has become one of the hot topics on the design and development of network access technologies. In particular, its authentication and key exchange (AKE) aspects, which form a vital building block for modern security mechanisms, deserve further investigation. In this paper we first identify the general requirements used for WLAN authentication and key exchange (AKE) methods, and then classify them into three levels (mandatory, recommended, and additional operational requirements). We present a review of issues and proposed solutions for AKE in 802.11 WLANs. Three types of existing methods for addressing AKE issues are identified, namely, the legacy, layered and access control-based AKE methods. Then, we compare these methods against the identified requirements. Based on the analysis, a multi-layer AKE framework is proposed, together with a set of design guidelines, which aims at a flexible, extensible and efficient security as well as easy deployment.
PDF [116.1 kB]
Method for carrying out a QoS-oriented handoff between a first and a second IP-based, especially mobile IPV6-based, communication path, between a mobile node (MN) and a correspondent node (CN) ,
Changpeng Fan, Andreas Festag, Xiaoming Fu , Cornelia Kapper, Holger Karl, Mirko Schramm, and Günter Schäfer (inventors), granted patent, No. AU2001276315,
August 2007.
Read abstract
The invention relates to a method for carrying out a QoS-oriented handoff between a first and a second IP-based, especially mobile IPv6-based, communication path, between a mobile node (MN) and a correspondent node (CN), the second communication path being part of a number of communication paths which can be accessed by the mobile node, with no, one, or a plurality of intermediate instances. The inventive method comprises at least the following steps: (a) a communication path is selected from the communication paths which can be accessed by the mobile node, as a second communication path; (b) a message (BU) is generated by the mobile node, said message containing at least one IP address which is associated with the mobile node on the basis of the selected communication path, and containing minimum quality of service requirements (QoS) in terms of the selected communication path; (c) the ability to meet at least the minimum quality of service requirements is controlled and optionally ensured by the individual intermediate instances through which the message passes successively, on the selected communication path and/or through the correspondent node. The message contains the minimum quality of service requirements for a communication from the mobile node to the correspondent node and/or vice versa. A handoff is automatically carried out between the first communication path and the second selected communication path, when at least the minimum quality of service requirements are met or the message is stopped. A notice is generated in an intermediate instance and/or in the correspondent node and is sent to the mobile node if the ability to meet the minimum quality of service requirements is not ensured.
Evaluating the Benefits of Introducing PMIPv6 for Localized Mobility Management ,
Jun Lei , and Xiaoming Fu , Technical Report No. IFI-TB-2007-02, Institute of Computer Science, University of Goettingen, Goettingen, Germany,
ISSN 1611-1044, June 2007.
Read abstract
Since recent years, it has been recognized that using global mobility protocol for managing localized mobility causes a number of problems, such as long registration delay. To overcome these problems, host-based and network-based localized mobility approaches have been proposed. Moreover, network based mobility management is more desirable since it requires no host software stack changes. Proxy Mobile IPv6 (PMIPv6) provides a solution for network-based mobility management that can avoid tunneling overhead over the air and support for hosts without an involvement in the mobility management.
We first review the localized mobility proposals and explore three major benefits that PMIPv6 can bring. In particular, we evaluate two aspects of the handover performance through a mathematical model for Fast Handovers for MIPv6 (FMIPv6), Hierarchical MIPv6 (HMIPv6), Fast handovers for HMIPv6 (F-HMIPv6) and PMIPv6. These analytical studies show that PMIPv6 may cause high handover latency if the local mobility anchor (LMA) is located far from the current mobility access gateway (MAG).
In this paper, we therefore propose an enhancement for PMIPv6, so-called fast handovers for PMIPv6 (F-PMIPv6) to further reduce the handover latency. The analysis result ascertains that F-PMIPv6 is a promising mobility scheme to efficiently manage the localized mobility.
PDF [184.3 kB]
ENABLE QoS Services for Large Operational IP Mobility Networks ,
Ivano Guardini, and Xiaoming Fu , Presented at the 1st OpenNet Workshop, Brussels, Belgium,
March 2007.
E2T: End-to-End Tunnelling Extension to Mobile IPv6 ,
Deguang Le , Xiaoming Fu , Xiaoyuan Gu, and Dieter Hogrefe, in Proceedings of the 4th IEEE Consumer Communications and Networking Conference (CCNC 2007), Las Vegas, Nevada, USA,
IEEE Communications Society, January 2007.
Read abstract
In the standard Mobile IPv6 (MIPv6), the bidirectional tunnelling through the home agent or the route optimization show inefficiency in per-packet routing, especially when both communicating endpoints are mobile. To be scalable and compatible, mobile devices packets should be routed efficiently with minimal changes to the network infrastructure. However, the current solutions do not provide any means for the end systems to perform optimized packet routing during the operation of the mobile devices. In this paper, we present an end-to-end tunnelling extension to MIPv6 (E2T) for mobile routing packets, which reduces the per-packet routing cost for the communications of mobile devices through the lower packet routing overhead. Besides, our approach requires little change to MIPv6, but allows the more efficient routing behavior with the shorter end-to-end transmission latency between communicating endpoints. The simulation results show our approach is suitable for real-time multimedia applications.
PDF [357.5 kB]
DMMP: A New Dynamic Mesh-based Overlay Multicast Protocol Framework ,
Jun Lei , Xiaoming Fu , and Dieter Hogrefe, Proceedings of the 2007 IEEE Consumer Communications and Networking Conference - Workshop on Peer-to-Peer Multicasting (P2PM 2007), Las Vegas, Nevada, USA,
IEEE Communications Society, January 2007.
Read abstract
Multicasting can provide an efficient way of delivering data from a sender to a group of receivers. It has received much attention over the past decade because of an increasing demand for group communication applications such as multimedia streaming. However, native IP multicast has not become widespread largely due to its technical and operational issues. To overcome these obstacles of deployment, various application layer and overlay multicast approaches have been proposed. Compared with IP multicast, they provide a new way of handling multicast without upgrading the infrastructure in a large scale. Nevertheless, they introduce a number of challenges and are still plagued with concerns on scalability, heterogeneity and dynamic performance. In this paper we propose a new protocol framework for addressing these issues, so-called the Dynamic Mesh-based Overlay Multicast Protocol or DMMP, which intends to provide an efficient and resilient multicast support by dynamically managing an overlay core comprised of end hosts. Moreover, DMMP can be used for media streaming which is contracted by a limited resource in stream supplying entities and requires good scalability and reliability. Initial analysis shows that DMMP has the potential to efficiently deliver multicast services for large groups.
PDF [311.3 kB]
Probe-Aided MulTCP: An Aggregate Congestion Control Mechanism ,
Fang-Chun Kuo , and Xiaoming Fu , Technical Report No. IFI-TB-2007-01, Institute of Computer Science, University of Goettingen, Goettingen, Germany,
ISSN 1611-1044, 2007.
Read abstract
A number of new application scenarios, e.g. mobile/wireless TCP proxies, edge to edge overlays, QoS provisioning and mass data transport, are calling for aggregate flow management. In this paper we show that applying a single flowshare to an aggregate flow will result in unfairness in the bandwidth sharing between the aggregate traffic and the background flows sharing the same bottleneck. To overcome this problem, we propose an aggregate congestion control mechanism, namely probe-aided MulTCP, which dynamically adjusts the congestion window loop to support multiple flowshares for an aggregate. The probe-aided MulTCP differs from existing works, such as MPAT, CP, MulTCP, in the following aspects. Firstly, our simulations show that against the traditional MulTCP the probe-aided MulTCP could maintain relatively stable, smooth and fair performance over a wide range of weight N in steady conditions as well as in varied congestion conditions. Secondly, an adjusting window loop is introduced to constantly probe the congestion situation and dynamically adjust the congestion window size for the newly arriving and leaving flows within the aggregate. This integration of congestion information improves the startup performance for new arriving flows, especially for short-lived ones. Thus, the probe-aided MulTCP is lightweight since only one extra probe window loop is used. Our extensive simulation studies show that with the probe-aided MulTCP, the improved performance and fairness will overweight the complexity caused by two congestion window loops.
PDF [665.8 kB]
2006
MobiArch'06 - Proceedings of the First ACM/IEEE International Workshop on Mobility in the Evolving Internet Architecture ,
Katherine Guo, Xiaoming Fu , and Jon Crowcroft (editors), San Francisco, CA, USA,
ACM Press, ISBN 1-59593-566-5, December 2006.
Beyond QoS Signaling: a Generic IP Signaling Framework ,
Xiaoming Fu , Hannes Tschofenig , and Dieter Hogrefe, Computer Networks, Volume 50, Issue 17, pages 3416-3433,
Elsevier, December 2006.
Read abstract
This paper describes the design principles and an introduction of a framework and protocols for generic IP signaling, namely the Cross-Application Signaling Protocol (CASP) and its signaling applications. While reusing certain features of the existing RSVP protocol, CASP overcomes its shortcomings and may be deployed as a replacement technology to provide simpler, mobility-supported, more extensible and more secure signaling services in IP based networks. This paper discusses challenges of todays IP signaling protocols and addresses fundamentals and key aspects of CASP and its current signaling applications. In addition, a comparison with previous signaling protocol proposals and an outlook of future work in this area are also given.
PDF [602.3 kB]
From Resource Reservation to Extensible IP Signaling ,
Xiaoming Fu , Habilitation Thesis, Mathematische Fakultaet, Universitaet Goettingen, Goettingen, Germany, 146 pages,
October 2006.
Dynamic Mesh-based overlay Multicast Protocol (DMMP) ,
Jun Lei , Xiaoming Fu , and Dieter Hogrefe, Internet Research Task Force, Internet draft (draft-lei-samrg-dmmp-01), work in progress, Scalable Adaptive Multicast (SAM) Research Group,
October 2006.
Read abstract
This document describes a Dynamic Mesh-based overlay Multicast Protocol (DMMP) to support multicast data delivery applications without relying on classic IP multicast, including multicast group management, overlay hierarchy establishment, multicast tree construction and data forwarding scheme from the source to a number of receivers. The DMMP framework builds on control plane functions which dynamically manage an overlay core and a multicast tree layer. The key idea is a number of end hosts self-organize into an overlay mesh, and dynamically maintain such a mesh. Based on the constructed mesh, some core-based clusters are built with capacity-aware trees inside. Then, a multicast tree consisting of DMMP-aware end hosts (and/or specific routers) is built on the top of the overlay core for the efficient delivery of the multicast data.
PDF [64.0 kB]
Securing the Next Steps in Signalling (NSIS) Protocol Suite ,
Hannes Tschofenig , and Xiaoming Fu , International Journal of Internet Protocol Technology, Volume 1, No. 4, pages 271-282,
InderScience Publishers, ISSN 1743-8209, August 2006.
Read abstract
The Next Steps In Signalling (NSIS) protocol suite represents an extensible framework for enabling various signalling applications over IP-based networks. The framework consists of two layers that need different types of security protection; the lower layer mainly deals with the discovery of adjacent peers and establishment of channel security to protect the delivery of signalling messages between two peers, while the upper layer provides the signalling application specific functionalities. Different security properties are required at the two layers with stronger authorisation functionality at the signalling application layer. In this paper we examine how various security vulnerabilities can be utilised by an adversary, including eavesdropping, Man-In-The-Middle (MITM) attacks, fraud and Denial of Service (DoS) attacks. Moreover, we describe how to protect against a number of selected security threats and highlight some security challenges that require further research.
PDF [431.5 kB]
Implementation and Performance Study of a New NAT/Firewall Signaling Protocol ,
Niklas Steinleitner , Henning Peters , Xiaoming Fu , and Hannes Tschofenig , in Proceedings of the 26th International Conference on Distributed Computing Systems-Workshops (ICDCSW 2006), the 5th International Workshop on Assurance in Distributed Systems and Networks (ADSN2006), Lisboa, Portugal,
IEEE Computer Society, ISBN 0-7695-2541-5, July 2006.
Read abstract
The NAT/Firewall NSIS Signaling Layer Protocol (NAT/FW NSLP) is a path-coupled signaling protocol for explicit Network Address Translator and firewall configuration within an extensible IP signaling framework currently being developed by the IETF Next Steps in Signaling (NSIS) working group. This new protocol allows end hosts to signal along a path to configure NATs and firewalls according to the data flow needs. In this paper we present a first open source implementation and performance evaluation of the NAT/FW NSLP protocol. The implementation utilizes a generic state machine template and can automatically generate source code for message handling classes. The performance study shows that our implementation scales well and is able to support firewall signaling for up to tens of thousands of flows in parallel even in a low-end PC testbed environment. The overall performance bottleneck is found to lie in the utilized firewall implementation, not depending on the NAT/FW NSLP implementation.
PDF [394.6 kB]
DMMP: A New Dynamic Mesh-based Overlay Multicast Protocol Framework ,
Jun Lei , Xiaoming Fu , and Dieter Hogrefe, Technical Report No. IFI-TB-2006-05, Institute of Computer Science, University of Goettingen, Germany,
ISSN 1611-1044, July 2006.
Read abstract
Multicasting provides an efficient way of delivering data from a sender to a group of receivers. It has been gained much attention over the past decade because of an increasing demand for group communication applications such as multimedia streaming. Compared with network layer multicast solutions, recent application-layer multicast and overlay multicast approaches provide a new way of handling multicast without upgrading the infrastructure in a large scale. Meanwhile, they introduce a number of challenges and are still plagued with concerns pertaining to scalability, deployment, heterogeneity and dynamic performance. In this paper we propose a new protocol framework for relieving these issues, so-called the Dynamic Mesh-based Overlay Multicast Protocol or DMMP, which intends to provide an efficient and reliable multicast support by dynamically managing an overlay core comprised of end hosts. Although more analysis and evaluation is necessary, this paper sheds light on several identified design issues with DMMP and initially analyzes its performance.
PDF [689.2 kB]
Modelling Soft-State Protocols with SDL ,
Xiaoming Fu , IEE Proceedings Communications,
ISSN 1350-2425, Volume 153, Issue 3, pages 365-375, June 2006.
Read abstract
The notion of soft state has been introduced in packet-switched networks to achieve particular services for end-to-end communications, such as quality-of-service provisioning and configuration of stateful packet filters. Protocols built upon soft state principles were believed to be simple, however in practice they are far more complex. An important issue with such protocols is to ensure their operations to be error-free and deadlock-free. In the paper the use of formal techniques is proposed, specifically, Specification and Description Language (SDL) and Message Sequence Charts (MSCs), for modelling, analysis and validation of soft-state protocols. Based on a general state management system that identifies their most representative behaviour, an extensive study on modelling and validating soft-state protocols with SDL/MSCs is presented, and it is shown that design flaws and ambiguity introduced in informally specified, textual protocols can be avoided if a protocol is formally modelled.
PDF [341.0 kB]
GONE: an Infrastructure Overlay for Resilient, DoS-Limiting Networking ,
Xiaoming Fu , and Jon Crowcroft, Proceedings of the 16th ACM International Workshop on Network and Operating Systems Support for Digital Audio and Video (NOSSDAV 2006), Newport, Rhode Island, USA,
ACM, May 2006.
Read abstract
With today's penetration in volume and variety of information flowing across the Internet, data and services are experiencing various issues with the TCP/IP infrastructure, most notably availability, reliability and mobility. Therefore, a critical infrastructure is highly desireable, in particular for multimedia streaming applications. So far the proposed approaches have focused on applying application-layer routing and path monitoring for reliability and on enforcing stateful packet filters in hosts or network to protect against Denial of Service (DoS) attacks. Each of them solves its own aspect of the problem, trading scalability for availability and reliability among a relatively small set of nodes, yet there is no single overall solution available which addresses these issues in a large scale.
We propose an alternative overlay network architecture by introducing a set of generic functions in network edges and end hosts. We conjecture that the network edge constitutes a major source of DoS, resilience and mobility issues to the network, and propose a new solution to this problem, namely the General Internet Signaling Transport (GIST) Overlay Networking Extension, or GONE. The basic idea of GONE is to create a half-permanent overlay mesh consisting of GONE-enabled edge routers, which employs capability-based DoS prevention and forwards end-to-end user traffic using the GIST messaging associations. GONE's use of GIST on top of SCTP allows multi-homing, multi-streaming and partial reliability, while only a limited overhead for maintaining the messaging association is introduced. In addition, upon the services provided by GONE overlays, hosts are identified by their unique host identities independent of their topologies location, and simply require (de-) multiplexing instead of the traditional connection management and other complex functionality in the transport layer. As a result, this approach offers a number of advantages for upper layer end-to-end applications, including intrinsic provisioning of resilience and DoS prevention in a dynamic and nomadic environment.
PDF [283.3 kB]
Overhead and Performance Study of the General Internet Signaling Transport (GIST) Protocol ,
Xiaoming Fu , Henning Schulzrinne, Hannes Tschofenig , Christian Dickmann , and Dieter Hogrefe, IEEE INFOCOM 2006, Bacelona, Spain,
IEEE, April 2006.
Read abstract
The General Internet Signaling Transport (GIST) protocol is currently being developed as the base protocol component in the IETF Next Steps In Signaling (NSIS) protocol stack to support a variety of signaling applications. In this paper we present our study on the protocol overhead and performance aspects of GIST. We quantify network-layer protocol overhead and observe the effects of enhanced modularity and security in GIST. We developed a first open source GIST implementation at the University of Göttingen, and study its performance in a Linux testbed. A GIST node serving 45,000 signaling sessions is found to consume small amounts of CPU and memory (on average 1.1ms for processing a signaling message and 2.4KB memory for a session). Individual routines in the GIST code are instrumented to obtain a detailed profile of their contributions to the overall system processing. Important factors in determining performance, such as the number of sessions, state management, refresh frequency, timer management and signaling message size are further discussed. We investigate several mechanisms to improve GIST performance so as to be comparable with an RSVP implementation.
PDF [181.9 kB]
Comparison Studies between Pre-Shared and Public Key Exchange Mechanisms for Transport Layer Security ,
Fang-Chun Kuo , Hannes Tschofenig , Fabian Meyer , and Xiaoming Fu , Proceedings of the 9th IEEE Global Internet Symposium, in conjunction with IEEE INFOCOM 2006, Barcelona, Spain, pages 77-82,
IEEE, ISBN 3-937201-01-7, April 2006.
Read abstract
The pre-shared key based mechanisms for Transport Layer Security (TLS) were recently standardized by the IETF to extend the set of ciphersuites by utilizing existing key management infrastructures. The benefit of pre shared based mechanisms is the avoidance or reduction of the cryptographic operations used in public-key based mechanisms. However, so far there are no performance measurements for pre-shared key based ciphersuites available. In this paper, we present a systematic analysis and performance comparison between the pre-shared key exchange mechanisms and the standard public key exchange mechanisms in TLS. Our performance metrics are processing
time and transmitted amount of data for a handshake establishment. Furthermore, the interaction between the overall TLS handshake duration and the network environment is evaluated. The results for different key exchange mechanisms are comparatively studied and the design choices of pre-shared key based key exchange mechanisms have been validated. Experimental results give details about the performance improvement of the preshared key based mechanisms compared to the standard public key based mechanisms.
PDF [331.4 kB]
Principles and Experiments of Explicit Delay Control ,
Xiaoyuan Gu, Dirk Markwardt, Lars Wolf, and Xiaoming Fu , Proceedings of IEEE Consumer Communications and Networking Conference (CCNC'06), Las Vegas, USA,
IEEE, January 2006.
Read abstract
Real-time interactive multimedia applications are highly delay-sensitive, and packets that are out of delay boundaries are usually obsolete. However the current Internet Protocol lacks a way to control the lifetime of the packets explicitly. We propose a packet lifetime control mechanism called Explicit Delay Control (EDC) that embeds a Maximum Tolerable Delay (MTD) field in an IPv4 option. At each network node, the MTD is deducted by the singlehop delay. Packets that expire their lifetime are discarded and non-congestion related delay losses are signaled to the sender to reduce inaccuracy in delay estimations and to adapt to path changes. We implemented EDC in the Linux kernel. Our evaluation has shown that EDC is an effective scheme to ensure the legality of the packets, reduce the waste of bandwidth and processing time in the networks, and alleviate congestions.
PDF [207.6 kB]
Comparison Studies between Pre-Shared Key and Public Key Exchange Mechanisms for Transport Layer Security (TLS) ,
Fang-Chun Kuo , Hannes Tschofenig , Fabian Meyer , and Xiaoming Fu , Technical Report No. IFI-TB-2006-01, Institute of Computer Science, University of Göttingen, Göttingen, Germany,
ISSN 1611-1044, 2006.
Read abstract
The public-key based handshake process of TLS is regarded as part of bottleneck that significantly degrades the performance. The pre-shared key based key exchange mechanisms for TLS were recently standardized by the IETF for avoiding or reducing the cryptographic operations in public-key based mechanisms. However, so far there is no performance measurement for pre-shared key based key exchange suites available. In this paper, we present a systematic analysis of performance comparison between the pre-shared key exchange mechanisms and the standard public key exchange mechanisms in TLS. Our performance metrics are the processing time in both slow and fast processor machines as well as the transmitted data amount for a handshake establishment. Furthermore, the interaction of the overall TLS handshake duration and the network environment is evaluated. The results for different key exchange mechanisms are comparatively studied and the design choices of pre-shared key based key exchange mechanisms have been validated. It has been observed that pre-shared key based mechanisms perform better than the standard public key based mechanisms.
PDF [357.6 kB]
A Review of Mobility Support Paradigms for the Internet ,
Deguang Le , Xiaoming Fu , and Dieter Hogrefe, IEEE Communications Surveys and Tutorials, Volume 8, No. 1, First Quarter, pages 38-51,
IEEE, ISSN 1553-877X, 2006.
Read abstract
With the development of mobile communications and Internet technology, there is a strong need to provide connectivity for roaming devices to continuously communicate with other devices on the Internet at any time and anywhere. The key issue of this vision is how to support mobility in TCP/IP networks. In this paper, we review the TCP/IP protocol stack and analyze the problems associated with it in the mobile environment. We then investigate the mobility support techniques and existing solutions for providing mobility support on the Internet. We classify the proposed solutions based on the protocol layers and present paradigms for each category of layer. We also provide a comparison of the different solutions belonging to different categories, including their advantages and disadvantages. Results have shown that there is no single solution that perfectly addresses mobility support for the Internet. Finally, we conclude this survey with a recommendation of features that ought to be met in Internet mobility support.
PDF [235.5 kB]
2005
Architectural Thoughts and Requirements Considerations on Video Streaming over the Internet ,
Jun Lei , Ingo Juchem , Xiaoming Fu , and Dieter Hogrefe, Technical Report No. IFI-TB-2005-06, Institute of Computer Science, University of Goettingen, Germany,
ISSN 1611-1044, November 2005.
Read abstract
With increasing demands of multimedia information over the Internet, video streaming has been received explosive attentions. With respect to the real-time nature of video streaming, instable bandwidth, latency, noise, packet loss, retransmission and out of order packet delivery are all problems that can affect video streaming over the Internet. However, the traditional Internet traffic is not sensitive to these problems. Based on the general video streaming architecture, we give out some considerations on design and architectural mechanisms, namely, media server, media compression, media QoS control, media distribution services, media security mechanisms and protocol stacks for video streaming. For each of these areas, we present some existing methods and implementations. Then we propose architecture via overlay multicast integrated with proxy caching to achieve efficiency, flexibility and scalability. Finally, we conclude this issue and point out the research direction.
PDF [503.1 kB]
A Quality-of-Service Resource Allocation Client for CASP ,
Henning Schulzrinne, Hannes Tschofenig , Xiaoming Fu , and Jochen Eisl, Technical Report No. TB-IFI-2005-07, Institute of Computer Science, University of Goettingen, Germany,
ISSN 1611-1044, November 2005.
Read abstract
Signaling resource reservations is one of the possible applications of the Cross-Application Signaling Protocol (CASP). This document describes a client protocol that supports per-flow resource reservationin both sender- and receiver-directed modes operation.
PDF [99.8 kB]
Security Implications of the Session Identifier ,
Hannes Tschofenig , Henning Schulzrinne, Robert Hancock, Andrew McDonald, and Xiaoming Fu , Technical Report No. TB-IFI-2005-08, Institute of Computer Science, University of Goettingen, Germany,
ISSN 1611-1044, November 2005.
Read abstract
As one result of the analysis activities in the NSIS group it was realized that mobility and the ability to change the flow identifier causes problems with existing QoS reservations. To be able to associate a signaling message with existing state an identifier other than the flow identifier had to be used. Such an abstraction is achieved with the session identifier which allows identification of established state independently of the flow characteristics.
Although the introduction of a session identifier sounds simple and beneficial, it introduces a problem which is subsequently referred to as the session ownership problem.
This document describes the session ownership problem, the implications for an NSIS protocol and summarizes already discussed solutions.
PDF [79.7 kB]
NSIS: A New Extensible IP Signaling Protocol Suite ,
Xiaoming Fu , Henning Schulzrinne, Attila Bader, Dieter Hogrefe, Cornelia Kappler, Georgios Karagiannis, Hannes Tschofenig , and Sven Van den Bosch, IEEE Communications Magazine, Internet Technology Series, 43(10): 133-141,
IEEE, October 2005.
Read abstract
In the last few years, a number of applications have emerged that can benefit from network-layer signaling, i.e., the installation, maintenance and removal of control state in network elements. These applications include path-coupled and path-decoupled quality of service (QoS) management and resource allocation, as well as network debugging, NAT and firewall control. These applications call for an extensible and securable signaling protocol. This paper discusses some of the recent standardization efforts in the IETF for a new extensible IP signaling protocol suite (NSIS). We describe the design of the NSIS protocol suite, and compare them with RSVP, the current Internet QoS signaling protocol.
PDF [159.9 kB]
Towards Self-optimizing Protocol Stack for Autonomic Communication: Initial Experience ,
Xiaoyuan Gu, Xiaoming Fu , Hannes Tschofenig , and Lars Wolf, In: Ioannis Stavrakakis and Michael Smirnov (eds), Proceedings of 2nd IFIP International Workshop on Autonomic Communication (WAC 2005), Athens, Greece, Lecture Notes in Computer Science, Volume 3854, pages 186-201,
Springer-Verlag, October 2005.
Read abstract
The Internet is facing ever-increasing complexity in the construction, configuration and management of heterogeneous networks. New communication paradigms are undermining its original design principles. The mobile Internet demands a level of optimum that is hard to achieve with a strictly-layered protocol stack. Questioning if layering is still an adequate foundation for autonomic protocol stack design, we study the state-of-the-art from both the layered camp and its counterpart. We then outline our vision on protocol stack design for autonomic communication with the POEM model and its internals. A novel cross-layer design approach that combines the advantages of layering and the benefits of holistic and systematic cross-layer optimization is at the core of this work. With inspirations from the natural ecosystem, we are working on the role-based Composable Functional System for self-optimization that features proactive monitoring and control. By doing so step-by-step, we envisage reaching the goal of self-tuning autonomic network with high level of autonomy and efficiency, with minimum human management complexity and user intervention.
PDF [296.8 kB]
Fast Seamless Handover Scheme and Cost Performance Optimization for Ping-Pong Type of Movement ,
Zongkai Yang, Yuming Wang, Dasheng Zhao, Jianhua He, and Xiaoming Fu , Proceedings of the 16th Annual IEEE International Symposium on Personal Indoor and Mobile Radio Communications (PIMRC 2005), Berlin, Germany,
IEEE, September 2005.
Read abstract
The ping-pong type of movement is a typical motion manner in mobile IPv6 networks, which will bring frequent handovers and thus increase signaling burden. On the other hand, reducing handover delay in this case seems to be more significant. In this paper we propose a fast seamless handover scheme for the ping-pong type of movement as an extension to the hierarchical mobile IPv6. Based on the simulation results, it can be observed that, by setting the reservation active flag (RAF) and the offline count down timer (CDT), the scheme significantly reduces QoS signaling cost and handover delay. Furthermore, the simulations work out an optimized CDT for acquiring better cost performance of resource reservation.
PDF [286.8 kB]
Advanced Authentication and Authorization for Quality of Service Signaling ,
Tseno Tsenov, Hannes Tschofenig , Xiaoming Fu , and Eckhart Koerner, 1st IEEE Workshop on Security and QoS in Communication Networks (SecQoS 2005), Athens, Greece (in conjunction with the first IEEE International Conference on Security and Privacy for Emerging Areas in Communication and Networks - SECURECOM 2005), Pages 224-235,
IEEE Computer Society Press, September 2005.
Read abstract
One of the key requirements of todays and future network infrastructures is to provide Quality of Service (QoS) support for end-to-end applications, by distinguishing the application flows and properly handling them in network nodes. As an important component to achieve Internet QoS, explicit signaling schemes for resource reservation have been proposed, which deal with admission, installation and refreshment of QoS reservation state information. To be useful, any QoS signaling protocol should provide a capability for authentication and authorization of the QoS requests, especially in environments where the end points are not trusted by the network nodes. However, existing protocols for QoS signaling encounter a number of authentication and authorization issues, which limit their application scenarios. The advent of NSIS QoS Signaling Layer Protocol (QoS-NSLP) offers the prospect to overcome some of these issues. After describing the overall design of QoSNSLP, we present an approach to support advanced authentication and authorization capabilities by using the Extensible Authentication Protocol (EAP). In comparison with existing approaches, this approach, combined with the support for effective interaction with the Authentication, Authorization and Accounting (AAA) infrastructure, provides flexible and extensible authentication and authorization methods for the QoS signaling.
PDF [959.2 kB]
E2T: End-to-End Tunneling Extension to Mobile IPv6 ,
Deguang Le , Xiaoming Fu , Xiaoyuan Gu, and Dieter Hogrefe, Technical Report No. IFI-TB-2005-05, Institute of Computer Science, University of Göttingen, Göttingen, Germany,
ISSN 1611-1044, September 2005.
Read abstract
In the standard Mobile IPv6, route optimization or bidirectional tunnelling through the home agent show inefficiency in per-packet forwarding, especially when both communicating endpoints are mobile. To be scalable and compatible, mobile devices packets should be forwarded in a way with minimal changes to the network infrastructure. However, the current solutions do not provide any means for the end systems to perform optimized packet routing during the operation of mobile devices.
In this paper, following a performance analysis of Mobile IPv6 routing mechanisms, we present the E2T - an extension to Mobile IPv6 for routing packets. It reduces per-packet forwarding cost for the communications of mobile devices. With this approach, packets are routed thorough end-to-end tunnelling between communicating endpoints, which requires little change to Mobile IPv6, but allows more efficient forwarding behavior. The numerical analysis and simulation results show it requires less overhead than the standard route optimization and it helps to achieve a low end-to-end traffic delay.
PDF [323.6 kB]
Enabling Mobile IPv6 in Operational Environments ,
Xiaoming Fu , Hannes Tschofenig , Srinath Thiruvengadam, and Wenbing Yao, in: Pascal Lorenz (ed), Proceedings of the 10th IFIP International Conference on Personal Wireless Communications (PWC 2005), Colmar, France, pp. 365-372,
Imperial College Press, ISBN 1-86094-582-1, August 2005.
Read abstract
Although Mobile IPv6 allows maintaining transport layer connections alive when an IPv6 node roams to different access networks, certain enabling mechanisms are needed for it to work in large scale network scenarios, including, most notably, issues with Mobile IPv6 bootstrapping and firewall traversal. This paper tries to address these problems by extending the IETF PANA and NSIS protocols to form an extensible framework for wide deployment of a secure, light-weight mobility service in operational IPv6 environments.
PDF [162.4 kB]
RSVP Standards Today and the Path Towards a Generic Messenger ,
Xiaoming Fu , and Jukka Manner, In: H. de Meer and N. Bhatti (eds.), Proceedings of the 13th International Workshop on Quality of Service (IWQoS 2005), Passau, Germany, Lecture Notes in Computer Science, Volume 3552, pages 385-387,
Springer Verlag, ISBN 3-540-26294-6, June 2005.
Read abstract
RSVP is a very well-known protocol to support resource reservations in IP-based networks. This paper provides a preliminary inventory of RSVP standards and discusses the path towards a generic messenger for Internet signaling.
PDF [62.4 kB]
Modeling Route Change in Soft State Signaling Protocols Using SDL: a Case of RSVP ,
Constantin Werner, Xiaoming Fu , and Dieter Hogrefe, in A. Prinz, R. Reed and J. Reed (eds.), Proceedings of the 12th SDL Forum (SDL 2005), Grimstad, Norway, Lecture Notes in Computer Science, Volume 3530, pages 174-186,
Springer Verlag, ISBN 3-540-26612-7, June 2005.
Read abstract
Soft state signaling protocols install and maintain states in network nodes, expiring without receiving refreshes. These states require proper reparation when the flow path changes, especially in case of link or node failures. As the specifications usually do not describe in detail how to handle these failures, we present insights by developing SDL models for RSVP on this issue.
PDF [269.3 kB]
Modeling Soft State Protocols with SDL ,
Xiaoming Fu , and Dieter Hogrefe, In: R. Boutaba et al. (Eds.), Proceedings of the 4th IFIP International Conference on Networking (Networking 2005), Waterloo, Canada, Lecture Notes in Computer Science, Volume 3462, pp. 289-302,
Springer Verlag, ISBN 3-540-25809-4, May 2005.
Read abstract
Soft state provides new services to packet-switching networks by introducing a type of state in the network nodes which is refreshed by periodical messages and otherwise expires. The operations of soft state protocols, which are being designed with ever greater complexity, need to be error-free and deadlock-free to avoid misusing network resources. Thus, verification, formal analysis and validation of these protocols become a vital task. In this paper we utilize formal techniques, specifically Specification and Description Language (SDL) and Message Sequence Charts (MSCs), for modeling, analysis and validation of various soft state protocols. We propose a general architecture for state management systems and find employing these techniques can help identify and correct possible design errors, which may be caused by informal specifications.
PDF [162.9 kB]
Analysis of Existing Quality-of-Service Signaling Protocols ,
Jukka Manner, and Xiaoming Fu , Request for Comment (RFC) 4094,
Internet Engineering Task Force, May 2005.
Read abstract
This document reviews some of the existing Quality of Service (QoS) signaling protocols for an IP network. The goal here is to learn from them and to avoid common misconceptions. Further, we need to avoid mistakes during the design and implementation of any new protocol in this area.
PDF [68.0 kB]
Performance Analysis of the TCP/IP Stack of Linux Kernel 2.6.9 ,
Jan Demter , Christian Dickmann , Henning Peters , Niklas Steinleitner , and Xiaoming Fu , Technical Report No. IFI-TB-2005-03, Institute of Computer Science, University of Göttingen, Germany,
ISSN 1611-1044, April 2005.
Read abstract
This document reports the project "performance study of the TCP/IP stack for the Linux kernel" which we performed during the practical course Computer Networks in winter semester 2004/05, including its design, implementation and performance results. We analysed the packet processing time traversing each layer of the Linux kernel 2.6.9 TCP/IP stack (socket, TCP/UDP, IP and Ethernet) and the influence of multi-threading and different packet sizes. The design is based on the idea of inserting probing points via hooks in the kernel code and export timing data to a userspace application. A packet generator and analysis tools were also developed. The results demonstrate a number of key concepts in TCP/IP networking, such as layering, user-system interface, connection versus datagram modes, processing routines and their overhead in different layers. Some preliminary results reveal the system has its bottlenecks in different situations, and our tools released under GPL-license have been designed in such a way that allows easy extensibility for other networking diagnostics purposes.
PDF [246.1 kB]
A Review of Mobility Support Paradigms for the Internet ,
Deguang Le , Xiaoming Fu , and Dieter Hogrefe, Technical Report No. IFI-TB-2005-01, Institute of Computer Science, University of Göttingen,
ISSN 1611-1044, January 2005.
Read abstract
With the development of mobile communication and Internet technology, there is a strong need to provide connectivity for roaming devices to communicate to other communication end points in the Internet at any time and anywhere. The key issue of this vision is how to support mobility in TCP/IP networks. In this paper, we review the TCP/IP protocol stack and analyze the problems associated with it in a mobile environment. We then investigate the mobility support techniques and existing solutions to provide mobility support in the Internet. We classify the proposed solutions based on the protocol layers and present examples for each category. We also provide a comparison of the different solutions belonging to different categories and in the same category, including their advantages and disadvantages, and conclude that there is no single solution perfectly addresses mobility support for the Internet.
PDF [347.0 kB]
2004
A Method for Authentication and Key Exchange for Seamless Inter-Domain Handovers ,
Rene Soltwisch, Xiaoming Fu , Dieter Hogrefe, and Sathya Narayanan, Proceedings of 12th IEEE International Conference on Networks (ICON 2004), Singapore, pp. 463-469,
ISBN 0-7803-8783-X, November 2004.
Read abstract
With the rapid growth of the Internet and mobile wireless technologies, an ever-increasing requirement on securing services between mobile users and access networks has become especially important. When a user roams into a foreign network, in addition to data confidentiality, mutual authentication between the user and the provider is also a vital issue. These concerns and the desire to stay seamlessly connected lead to the demand of fast authentication and key establishment mechanisms, which are particularly difficult in inter-domain handover scenarios. In this paper, we introduce a novel mechanism to provide a simple but effective method, which forwards the key from the previous access router to the new access router that the mobile node attaches to. With this mechanism, trust relationship can be re-established even if the access routers do not trust each other in such an inter-domain scenario. Compared with the classical authentication method used in GSM and a recently proposed EAP-based secure key exchange protocol, our approach shows advantages of faster key exchange and authentication with only minimal message exchange in the wireless link.
PDF [263.1 kB]
QoS and Security in 4G Networks ,
Xiaoming Fu , Dieter Hogrefe, Sathya Narayanan, and Rene Soltwisch, Proceedings of the 1st CIC/IEEE Global Mobile Congress (GMC 2004), Shanghai, China, pp. 117-122,
October 2004.
Read abstract
Future 4G mobile communication networks are expected to provide all IP-based services for heterogeneous wireless access technologies, assisted by mobile IP to provide seamless Internet access for mobile users. Two major challenges in developing such heterogeneous network infrastructure are QoS provisioning and security services for mobile users communication flows. This paper proposes a new architectural view and methodologies for QoS and security support in 4G networks, which integrates QoS signaling with authentication, authorization and accounting (AAA) services to both guarantee the user applications QoS requirements and achieve efficient authentication, authorization and key exchange.
PDF [369.9 kB]
Implementation and Evaluation of the Cross-Application Signaling Protocol (CASP) ,
Xiaoming Fu , Dieter Hogrefe, and Sebastian Willert , Proceedings of the 12th IEEE International Conference on Network Protocols (ICNP 2004), Berlin, Germany, pp. 61-71,
IEEE Computer Society Press, ISBN 0-7695-2161-4, October 2004.
Read abstract
In this paper, we describe implementation aspects and performance results of a novel general signaling protocol for the Internet, the Cross-Application Signaling Protocol (CASP). There has been much debate on the applicability of RSVP as a general signaling protocol for the Internet, particularly with respect to its modularity, complexity, security and mobility support. Based on a layered architecture, the CASP design intends to address these challenges, which, unlike RSVP, provides a simpler mechanism for reliability and security by re-using existing protocols for transporting signaling messages. In addition, it supports a wide range of signaling applications. While this concept is considered to be advantageous over RSVP signaling, the actual mechanisms and behaviors of the CASP implementation have not yet been explored. Our study attempts to shed light on this issue by presenting a first public CASP implementation and preliminary examination of its properties. Performance results show and analyze the round trip times and their variances of signaling messages upon different number of signaling requests and different congestion situations in the experimental setup. The memory required for a large number of signaling sessions and the CPU consumption for each routine from profiling the implementation are low. Although further work is necessary, critical design choices in CASP have been proven useful and practically feasible.
PDF [111.3 kB]
Modeling Soft State Protocols with SDL ,
Xiaoming Fu , and Dieter Hogrefe, Technical Report No. IFI-TB-2004-02, Institute of Computer Science, University of Göttingen, Germany,
ISSN 1611-1044, August 2004.
Read abstract
Soft state enables new services to packet-switching networks by introducing a type of state in the network nodes which is refreshed by periodical messages otherwise expire. System designers build protocols that implement soft state concepts based on intuition or on high-level explanations believe that the design is "better" than hard state and soft state implementations should be robust, reliable and interoperable. As states in the network nodes are critical for both applications the and network infrastructure, the operations of soft state protocols, which tend to be designed more and more complex, need to be error-free and deadlock-free. Thus, verification, formal analysis and validation of these protocols become a vital task. In this paper we utilize formal techniques, specifically, Specification and Description Language (SDL) and Message Sequence Chart (MSC), for modeling, analysis and validation of general soft state protocols. We propose a general architecture of state management systems and find several points through the SDL/MSC modeling which may enrich the design, modeling and evaluation of real soft state protocols: 1) modeling these protocols using these techniques is feasible, 2) it can be possible to use these techniques to identify possible design errors and deadlocks/livelocks, which may be caused by imprecise informal specifications of these protocols.
PDF [112.1 kB]
Implementation and Evaluation of the Cross-Application Signaling Protocol (CASP) ,
Xiaoming Fu , Dieter Hogrefe, and Sebastian Willert , Technical Report No. IFI-TB-2004-001, Institute of Computer Science, University of Göttingen, Germany,
ISSN 1611-1044, April 2004.
Read abstract
In this report, we describe implementation aspects and performance results of a novel general signaling protocol for the Internet, the Cross-Application Signaling Protocol CASP). Much debate exists about the applicability of RSVP as a general signaling protocol in the Internet, particularly for its modularity, complexity, security and mobility support. Based on a layered architecture, the CASP design intends to address these challenges and unlike RSVP, it provides a simpler mechanism for reliability and security by re-using existing protocols for transporting signaling messages, and supports a wide range of signaling applications. While this concept is considered to be advantageous over RSVP signaling, the actual mechanisms and behaviors of the CASP implementation have not yet been explored. With our work, despite being still far from a final judgment, we try to shed light on this issue by presenting a first public CASP implementation and a preliminary study about its properties. Performance results show that even under heavy signaling loads, the round trip time of signaling messages is acceptable (appr. 5ms in serving more than 1000 simultaneous signaling client applications in the initiator each at a random refresh interval between 3s and 15s in our experiments), and the memory and CPU consumption of the implementation are low. Although further work will be necessary, critical design choices in CASP have been proved to be feasible.
PDF [416.3 kB]
2003
Secure, QoS-Enabled Mobility Support in IP-based Networks ,
Xiaoming Fu , Tianwei Chen, Andreas Festag, Holger Karl, Günter Schäfer, and Changpeng Fan, Proceedings of the 4th Annual IP-based Cellular Network Conference (IPCN 2003), Paris, France,
December 2003.
Read abstract
The rising number of mobile users, the advent of various radio access technologies, and the increasing importance of IP services over wireless as well as wired networks pose a number of new challenges. While Mobile IP has been designed for mobility management in IP networks, it may result in high latency and signaling overhead during handoff. Thus, advanced mobility mechanisms improving Mobile IP are desired to perform efficient handoffs. Also, appropriate Quality-of-Service (QoS) support is needed for mobility-enhanced IP in order to meet end users expectations. Furthermore, security measures are required to protect the network infrastructure.
This paper describes the Secure, QoS-enabled Mobility (SeQoMo) architecture addressing these issues. In particular, optimization of handoff operations, low latency QoS re-establishment for IP-level handoff, authentication, and QoS-aware authorization for mobile nodes are investigated and integrated in a unified framework. We also describe how the SeQoMo architecture as a whole supports efficient handoff processing especially during local movements, with optimized QoS support and authentication and QoS-aware authorization services.
PDF [71.8 kB]
Prototype Implementation and Performance Evaluation of a QoS-Conditionalized Handoff Scheme for Mobile IPv6 Networks ,
Axel Neumann, Xiaoming Fu , and Holger Karl, Proceedings of the 18th Annual IEEE Computer Communications Workshop (CCW 2003), California, USA, pp. 24-29,
IEEE Press, ISBN 0-7803-8239-0, October 2003.
Read abstract
Future internetworks will include large numbers of portable devices moving among small, wireless cells. In order to support real-time applications, users demand seamless mobility and Quality-of-Service (QoS)provisioning. One approach towards a more flexible, customizable and scalable mobility architecture that also reduces signaling load and handoff latency results from the introduction of micro-mobility. Furthermore, by coupling QoS signaling and mobility management, QoS requirements can be negotiated without incurring significant additional signaling latency.
This paper presents the prototype implementation and performance evaluation of such a QoS-enabled micro-mobility scheme, which is called "QoS-conditionalized handoff". We extended the Mobile IPv6 for Linux implementation to support the basic mode of Hierarchical Mobile IPv6 as the underlying micro-mobility mechanism. One problem that appeared during the implementation was the rather complex event handling in the mobile node; to enable a simple and generic way of event handling, a priority-based execution structure has been developed that can be easily adapted to various policies.
Our experimental results show that by this QoS-conditionalized handoff scheme, QoS-enabled handoffs can be achieved with a small amount of introduced latency compared to Hierarchical Mobile IPv6, which is much less than that of Mobile IPv6. It is further observed that fewer packets were lost and registration latency could be much more decreased when mobility management in the mobile node takes advantage of a movement detection mechanism to expedite the QoS-conditionalized handoff procedure.
PDF [87.7 kB]
Development of QoS Signaling Protocols in the Internet ,
Xiaoming Fu , Proceedings of the 28th Annual IEEE Conferfence on Local Computer Networks (LCN 2003), Bonn/Königswinter, Germany, pp. 636-637, Workshop on High-Speed Local Networks,
IEEE Computer Society Press, ISBN 0-7695-2037-5, October 2003.
Read abstract
QoS signaling protocol is one of the key components in Internet QoS architectures to establish, maintain, and remove reservation states in network nodes. This paper gives an overview of the recent efforts underway on next steps in QoS signaling protocols, namely RSVP extensions with mobility support, QoS-conditionalized handoff protocol, the layered architecture RSVP Lite and the Cross-Application Signaling Protocol (CASP). These efforts address main issues with existing approaches differently : modularity, complexity and mobility support, with a focus on protocol behaviors based on different design principles. The paper also provides pointers to standards effort towards general Internet signaling and other service-specific signaling protocols.
PDF [39.6 kB]
Mobility Support for Next-Generation Internet Signaling Protocols ,
Xiaoming Fu , Henning Schulzrinne, and Hannes Tschofenig , Proceedings of the IEEE 58th Vehicular Technology Conference (VTC 2003-Fall), Orlando, Florida, USA, pp. 1979-1983, Symposium on IP Mobility,
IEEE, ISBN 0-7803-7954-3, October 2003.
Read abstract
Internet signaling protocols establish, maintain and remove state along the data path. Next-generation signaling protocols design must meet the scaling requirements imposed by the various tasks of the Internet signaling applications, such as resource reservation and middlebox configuration, and to meet the demand for general functionality in signaling protocols, including strong security, reliability, congestion control, support for various signaling purposes and message sizes, and efficient support for mobility. This paper presents a generic signaling architecture, the Cross-Application Signaling Protocol (CASP) and describes how it supports efficient and secure signaling in IP mobility scenarios. In this approach, the signaling functionality is splitted into two layers: a generic messaging layer which provides the generic functionality for message delivery, and a client layer consisting of a next-hop discovery client and any number of client protocols which perform the actual signaling tasks. The essential mechanisms required to support mobility are: (1) a session identifier uniquely selected by the initiator and effective discovery of the cross-over node; (2) a branch identifier incrementally assigned for the new branch and efficient release of state in the abandoned branch; (3) ensuring discovery messages are delivered exactly following the path that mobile IP packets are encapsulated; (4) effective hop-by-hop authentication and reauthorization provided by the messaging layer, non hop-by-hop security for signaling clients and denial-of-service protection in the discovery client.
PDF [75.9 kB]
CASP - Cross-Application Signaling Protocol ,
Henning Schulzrinne, Hannes Tschofenig , Xiaoming Fu , and Andrew McDonald, Technische Berichte des Instituts für Informatik an der Georg-August-Universität Göttingen,
Institut für Informatik, Georg-August-Universität Göttingen, Germany, ISSN 1611-1044, IFI-TB-2003-01, (equivalent to the Internet draft), March 2003.
Read abstract
CASP is a modular potocol for establishing network control state along a data path between two nodes communicating on the Internet.
The signalling problem addressed by CASP is the same as the overall problem being addressed by the NSIS activities.
The CASP framework is defined as a modular protocol, which includes a general purpose messaging layer (M-layer), which supports a number of client layers for particular ignalling applications (e.g. QoS, MIDCOM). In addition there is distinct, special purpose client component for next-peer discovery.
PDF [103.1 kB]
Towards RSVP Lite: Light-weight RSVP for Generic Signaling ,
Xiaoming Fu , and Cornelia Kappler, Proceedings of the 17th International Conferfence on Advanced Information Networking and Applications, Xi'an, China, pp. 619-622,
IEEE Computer Society Press, ISBN 0-7695-1906-7, March 2003.
Read abstract
RSVP is a reservation setup protocol designed specifically to support QoS signaling in the Internet. However, RSVP end-to-end signaled QoS for the Internet has not become a reality. Moreover, there are many other applications demanding different signaling services. This paper analyses the features of RSVP version 1 we believe to be essential, and its complexity due to QoS-oriented design and multicast support as an indispensable component in a signaling protocol, deriving the design principles to be covered in a more generic signaling protocol. Based on this analysis, we present a light-weight version of RSVP, RSVP Lite, which clearly separates the signaled data from signaling messages and removes the multicast capability from the mandatory components of RSVP. RSVP Lite is intended to be applicable to a wide range of networking environments, while providing the flexibility to serve for generic signaling purposes and incremental deployment in the Internet.
PDF [42.9 kB]
Design of CASP - a Technology Independent Lightweight Signaling Protocol ,
Henning Schulzrinne, Xiaoming Fu , Cornel Pampu, and Cornelia Kappler, Proceedings of the 1st International Workshop on Inter-domain Performance and Simulation (IPS 2003), Salzburg, Austria,
February 2003.
Read abstract
Existing signaling solutions are insufficient in terms of inter-domain and out-of-path signaling, mobility support and inter-working with policy and security mechanisms. The paper presents the Cross-Application Signaling Protocol (CASP) which is a general-purpose protocol for managing state information in network devices. This technology independent signaling protocol can be used for inter- and intra-domain QoS signaling, the configuration of middleboxes, for collecting measurement data and any other application where state management is required. It relies on existing transport protocols and consists of a messaging layer and a client layer. The messaging layer is application independent and is responsible for routing, session establishment and feature negotiation. In contrast to this application independent component of CASP, the client layer is the application-dependent part. As an example for a client the paper describes the QoS Resource Allocation Client for CASP and discusses requirements for extending CASP to include interdomain signaling. The discovery of next peers along the data path is handled by the Scout protocol, which is a specialized client protocol. Some of the basic mechanisms are derived from existing protocols. This way the design of this protocol relies on the experiences made in this area and is therefore one of the promising protocol candidates for the IETF NSIS WG.
PDF [48.2 kB]
2002
Analysis on RSVP Regarding Multicast ,
Xiaoming Fu , Cornelia Kappler, and Hannes Tschofenig , Technische Berichte des Instituts für Informatik an der Georg-August-Universität Göttingen,
Institut für Informatik, Georg-August-Universität Göttingen, Germany, ISSN 1611-1044, IFI-TB-2002-001, October 2002.
Read abstract
RSVP version 1 has been designed for optimum support multicast. However, in reality multicast is being used much less frequently than anticipated. Still, even for unicast (one sender, one receiver) full-fledged multicast-enabled RSVP signaling must be used. As pointed out in the NSIS requirement draft, multicast would not be necessarily required for an NSIS signaling protocol. This draft analyses ingredients of RSVP Version 1 which are affected by multicast, and derives how these ingredients may look like if multicast is not supported in the generic RSVP signaling protocol and adapt related functionalities accordingly - we call the resulting feature set "RSVP Lite", a potentially more light-weight version of RSVP.
PDF [335.7 kB]
QoS-Conditionalized Handoff for Mobile IPv6 ,
Xiaoming Fu , Holger Karl, and Cornelia Kappler, Proceedings of the 2nd IFIP International Networking Conference (NETWORKING 2002), Pisa, Italy, Lecture Notes in Computer Science, Volume 2345, pp.721-730,
Springer Verlag, ISBN 3-540-43709-6, May 2002.
Read abstract
In this paper we present a scheme that enables a mobile user to perform a "QoS-conditionalized" handoff when moving to an overlapping area in Mobile IPv6. The idea is to use a QoS hop-by-hop option piggybacked in the binding messages for QoS signaling and conditionalize a handoff upon the availability of sufficient resources along the new transmission path. Our scheme builds upon the hierarchical mobile IPv6 protocol and is especially suited for micro-mobility. It also enables the mobile node to flexibly choose among a set of available access points so that the mobile node can transmit packets through a route which offers satisfying QoS.
PDF [110.4 kB]
2001
PRM: A Resource Management Framework for Policy-driven QoS Control in Enhanced Internets ,
Xiaoming Fu , and Yaoxue Zhang, Chinese Journal of Electronics, 10(1): 13-18,
ISSN 1022-4653, October 2001.
Read abstract
The Internet has evolved from a traditional best-effort delivery data network into an enhanced Internet that can provide a certain Quality of Service (QoS) mechanisms for applications. However, these QoS mechanisms usually do not take policy factors into account such as priority and time-of-day. The paper presents a Resource Management Framework for Policy-driven QoS control (PRM) in enhanced internets, which consists of four components: Domain Policy Controller (DPC) resides an administrator domain, Border Resource Manager (BRM) resides between adjacent domains, Interior Resource Manager (IRM) in each router between two BRMs within the same domain, and End-system Resource Manager (ERM) in charge of end-node/host router resources. We first introduce challenges and demands facing the enhanced internets, then give a formalized definition of policy and describe the functions and interaction of components of PRM. Specifically we take QoS-pricing policy used in a DiffServ environment as an example to illustrate the operational phases of PRM.
PDF [99.9 kB]
2000
Admission Control for Providing Statistical QoS in High-Speed Networks ,
Xiaoming Fu , and Yaoxue Zhang, Acta Electronica Sinica, 28(10): 82-85,
October 2000.
Achieving QoS Request Efficiently for the Internet Using RSVP Tunnels ,
Guoqiang Guo, Yaoxue Zhang, and Xiaoming Fu , Journal of Computer Research and Development, 37(1):55-60,
, January 2000.
