Research
| I'm working as a Senior Technical Staff Member in the IP Network Management and Performance Department, Internet and Networking Systems Research Center, AT&T Labs Research. I graduated from Cornell University with a PhD in computer science under advice of Dr. Balachander Krishnamurthy (AT&T Labs -- Research, Florham Park, NJ) and Dr. Robbert van Renesse (Cornell University). My Ph.D. thesis is Network-Aware Client Clustering and its Applications [ps]. I did my M.S. work with Prof. Srinivasan Keshav. My research interests include network-aware clustering and its applications, BGP routing, Web performance, content distribution, DNS-based application, network dynamics and performance measurement, Internet computing, wireless network, and other Internet related research topic. |
 Network-aware
client clustering |
| 1. On Network-Aware Clustering of Web Clients [ps],
by Balachander Krishnamurthy and Jia Wang, Proceedings of ACM SIGCOMM'2000.
The full version of the paper is published as AT&T Labs--Research Technical
Memorandum HA1630000-000101-01TM [ps], January
1, 2000. A talk [ppt] is given at ACM
SIGCOMM'2000, Stockholm, Sweden, August 2000.
Being able to identify the groups of clients that are responsible for a significant portion of a Web site's requests can be helpful to both the Web site and the clients. In a Web application, it is beneficial to move content closer to groups of clients that are responsible for large subsets of requests to an origin server. We introduce clusters---a grouping of clients that are close together topologically and likely to be under common administrative control. We identify clusters using a "network-aware" method, based on information available from BGP routing table snapshots. Experimental results show that our entirely automated approach is able to identify clusters for 99.9% of the clients in a wide variety of Web server logs. Sampled validation results show that the identified clusters meet the proposed validation tests in over 90% of the cases. An efficient self-corrective mechanism increases the applicability and accuracy of our initial approach and makes it adaptive to network dynamics. In addition to being able to detect unusual access patterns made by spiders and (suspected) proxies, our proposed method is useful for content distribution and proxy positioning, and applicable to other problems such as server replication and network management. |
| 2. Fast Network-Aware
Clustering by Faster Prefix Matching, by Adam L. Buchsbaum, Glenn S. Fowler,
Balachander Krishnamurthy, and Jia Wang, submitted for publication. The
full version of the paper is published as AT&T Labs--Research Technical
Memorandum, 2000.
In prior work we introduced network-aware clustering as a way of identifying a set of IP addresses that are with high probability under common administrative control and topologically close together. A busy cluster is a cluster responsible for a significant subset of traffic to a Web site. Web site owners can benefit by knowing the identity of the busy clusters. In this work, we first introduce a novel data-structure --- retrie --- required for longest prefix matching to enable on-line clustering. The retrie data structure works well enough to outperform the best known longest prefix matching and hence is actually applicable for software routing. We also show experimentally that: (i) grouping clusters into super-clusters based on the originating AS information does not improve on prefix-level clustering (ii) non-busy clusters cannot be combined into a busy super-cluster and (iii) large clusters can be partitioned into sub-clusters based on the clients access pattern. Finally, we examine two new applications, server clustering and server mirroring, to demonstrate the use of our improved clustering scheme and the fast prefix matching algorithm. |
| 3. Network-Aware client clustering and applications,by Balachander Krishnamurthy and Jia Wang, preparing for submission. |
| Web
caching |
| 1. A Survey of Web Caching Schemes
for the Internet [ps], by Jia Wang, ACM Computer
Communication Review (CCR), Vol. 29, No. 5, October 1999. The full version
of the paper is published as Technical Report TR99-1747, Department of
Computer Science, Cornell University, May 13, 1999.
The World Wide Web can be considered as a large distributed information system that provides access to shared data objects. As one of the most popular applications currently running on the Internet, the World Wide Web is of an exponential growth in size, which results in network congestion and server overloading. Web caching has been recognized as one of the effective schemes to alleviate the service bottleneck and reduce the network traffic, thereby minimize the user access latency. In this paper, we first describe the elements of a Web caching system and its desirable properties. Then, we survey the state-of-art techniques which have been used in Web caching systems. Finally, we discuss the research frontier in Web caching. |
| 2. A Scalable Efficient Robust Adaptive Architecture
for the Next Generation of Web Services, by Jia Wang, Technical Report
TR99-1767, Department of Computer Science, Cornell University, September
14, 1999.
The World Wide Web can be considered as a large distributed information system that provides access to shared data objects. As one of the most popular applications currently running on the Internet, the World Wide Web is of an exponential growth in size, which results in network congestion and server overloading. Web caching has been recognized as one of the effective schemes to alleviate the service bottleneck and reduce the network traffic, thereby minimize the user access latency. In this paper, we propose a Scalable Efficient Robust Adaptive (SERA) Web caching architecture which accommodates the exponential growth and extreme dynamic environment of the World Wide Web. We developed a piggybacked prefetching/pre-resolving scheme, which uses user access pattern and network environment information. Cooperative consistency control mechanism is employed to further improve the performance. In order to assist the cache resolution, an efficient cache routing scheme is employed. A loose group membership is maintained among nearby proxy caches to make the entire caching system scalable and fault tolerant. |
| 3. Collaborative Web Caching: Architecture and Experiments, by Jiong Yang, Wei Wang, Richard Muntz, and Jia Wang, preparing for submission. |
| Network
performance measurement and modeling |
| 1. The Implication of Network Performance on Service
Quality: Testbed and Applications [ps], by Yu Zhang,
Jia Wang, and Srinivasan Keshav, submitted for publication. The full version
of the paper is available as Technical Report TR99-1754, Department of
Computer Science, Cornell University, June 29, 1999.
As the Internet infrastructure evolves to include Quality of Service (QoS), it is necessary to map application quality requirements to network performance specifications in terms of delay and loss rate. In this paper, we first propose a new testbed by using a combination of simulation and emulation technics. Next, we use this testbed in studying the effect of network performance metrics on the perceptual quality of various applications to demonstrate the applicability of the testbed. We examine two typical applications on the basis the new testbed: the World Wide Web and the Internet Telephony. In our study of the World Wide Web, we derived a new TCP short connection model that computes the latency of Web retrieval accurately and efficiently given only the packet delay and loss rate characteristics apriori. In the study of Internet telephony, we show that the packet delay variance is the dominant network characteristics which affect the perceptual quality. As a result, the service quality drops dramatically when the network offered load reaches 80%. This can serve as a guideline for studies towards improving service quality of Internet telephony. |
| 2. Efficient and Accurate Ethernet Simulation [ps],
by Jia Wang and Srinivasan Keshav, Proceedings of the 24th IEEE Annual
Conference on Local Computer Networks (LCN'99). The full version of the
paper is published as Technical Report TR99-1749, Department of Computer
Science, Cornell University, May 20, 1999.
The Internet is increasingly being called upon to provide different levels of service to different applications and users. A practical problem in doing so is that although Ethernet is one of the hops for nearly all communication in the Internet, it does not provide any QoS guarantees. A natural question, therefore, is the effect of offered load on Ethernet throughput and delay. In this paper, we present several techniques for accurately and efficiently modeling the behavior of a heavily loaded Ethernet link. We first present a distributed approach to exact simulation of Ethernet, which greatly simplifies collision detection. Then, we describe an efficient distributed simulation model, called Fast Ethernet Simulation, that empirically models an Ethernet link to quickly and accurately simulate it. By eliminating the implementation of CSMA/CD protocol, our approach reduces computational complexity drastically while still maintaining desirable accuracy. Performance results show that our techniques not only add very little overhead (less than 5% in our tests) to the basic cost of simulating an Ethernet link, but also closely match real-world measurements. We also present efficient techniques for compressing cumulative distributions using hyperbolic curves and for monitoring the load on a heavily-loaded link. |
| IP
Routing |
| 1. BGP Extension for Inter-Domain QoS Routing, by Jia Wang, Yu Zhang, and Srinivasan Keshav, preparing for submission. |
| 2. Performance Evaluation of the Protocol Independent (PI) Switching Set Under IP Traffic. This is the summer Intern project at Bell Labs, Lucent Technologies, under direction of Dr. Fabio Chiussi, Summer 1998. |
| Computer
Image Processing |
| 1. Image Disorientation Auto-Recovery, by Jia Wang, Yin
Zhang, Zhiyuan Chen, and Yu Zhang, Technical Report TR99-1729, Department
of Computer Science, Cornell University, Feb. 8, 1999.
Automatically detecting and correcting disoriented image frames in a content-related image sequence is a problem that must be addressed in many image and video applications. In this paper, we give a robust feature-based algorithm to efficiently detect and adjust the disorientation of images. |
| 2. Detecting Static Objects in Busy
Scenes, by Jia Wang and Wei-Tsang Ooi, Technical Report TR99-1730, Department
of Computer Science, Cornell University, Feb. 8, 1999.
Detecting static objects in scenes containing significant number of moving objects has several applications in video surveillance. One example is the detection of suspicious packages which is left unattended in an airport terminal or railway station. This paper outlines an approach to automatically detect static objects from a video sequence of a busy scene. Our approach consists of two phase : foreground object extraction and object matching. In the first phase, we find the foreground objects in current video frame, using an image of a background as reference. In the object matching phase, we try to match the objects with objects that appears before in previous frames. Matching is done based on three parameter : shape and position, intensity and edge. Temporary occluded of objects is also handled. We built a system based on our approach. Preliminary experiments shows that our system are able to identify static objects in a busy scene in real time. |
| Yesterday is a memory,
Tomorrow is a dream, Today is the reality, Make the most of it. |
 |
Feburary 18, 2001