Title: Coordinated En-Route Web Caching
Speaker: Xueyan TANG


Web caching is an important technique for reducing Internet access latency,
network traffic, and server load. To obtain the full benefits of caching,
multiple web caches are often deployed in the network. They cooperate with
each other in handling client requests. The overall caching performance
depends on how the cache contents are managed. In this talk, we propose a
novel coordinated en-route caching scheme for geographically distributed web
caches. The proposed scheme integrates both object placement and replacement
strategies, and dynamically makes caching decisions for different caches in
a coordinated fashion. It takes into consideration the status of candidate
caches, the access frequency and size of the object, and the network distances
between caches. We formulate the object placement problem as an optimization
problem and compute the optimal locations for caching objects using a low-cost
dynamic programming algorithm. In addition to analytical modeling, our study
is also accompanied by extensive simulation experiments based on both
synthetic workload and real traces from web caches. The results show that
the proposed scheme significantly outperforms existing algorithms that
consider either object placement or replacement at individual caches only.


Xueyan Tang received his BEng degree in computer science and engineering
from Shanghai Jiao Tong University, Shanghai, China in 1998. He is currently
a PhD candidate in the Department of Computer Science at the Hong Kong
University of Science and Technology. His research interests include web
content delivery, mobile computing, Internet technologies, and distributed
systems. See http://www.cs.ust.hk/~tangxy/ for further information.