S. Jin and A. Bestavros, "Popularity-Aware GreedyDual-Size Web Proxy Caching Algorithms," Proc. 20th Int'l Conf. Distributed Computing Systems (ICDCS '00), Apr. 2000.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....the disk cache and interval caching for the memory cache. Sen. et al. propose pre x caching[17] in which a proxy server stores the initial frames of popular clips. Other selective partial caching techniques [19] are also employed to help performing smoothing at proxy servers. Finally, work in [11] suggests taking into consideration of popularity information in cache replacement algorithms. Our work borrows the idea of using popularity information and extends the idea to streaming servers. Moreover, we allow popularity information to be speci ed statically or inferred dynamically. ....

....pro le will be fairly accurate. The di erence of the other two on line methods lies in how popularity is de ned. In cumulative, an object s popularity is de ned as the cumulative references on the object. Other research [16, 12] uses this mechanism when calculating object popularity. Although [11] uses a decay function to de emphasize the signi cance of past accesses over the long term, there is no fundamental di erence for a short term run. The alternative way of calculating object popularity, on the other hand, is to count only the number of active streams on an object. This is ....

S. Jin and A. Bestavros. Popularity-aware greedy dual-size web proxy caching algorithms. In Proceedings of ICDCS, June 2000.

....differences, namely variable size, variable cost and the more complex statistics of request patterns, preclude an easy transfer of caching techniques developed earlier for computer system memory. A large number of studies have focused on the design of efficient replacement policies, e.g. see [8] [9] [10] 11] and references therein for a sample literature. Proposed policies typically exploit either access recency (e.g. the LRU policy) or access frequency (e.g. the LFU policy) or a combination thereof (e.g. the hybrid LRFU policy) The numerous policies which have been proposed are often ....

.... C 0 ) Thus, the byte hit rate associated with the policy C 0 is simply given by BHR(C M j=1 p(#(j) s(#(j) 21) Another interesting observation is the relation of the optimal 0 to the well established Greedy Dual and Greedy Dual Size replacement policies described in [8] and [9]. Let c GD : be an arbitrary cost used by the Greedy Dual policies. Under optional eviction, the Greedy Dual policies in case of a cache miss prescribe ) 22) L # where L is a contribution of the temporal locality of reference to the replacement policy and # 0 is a weight ....

S. Jin and A. Bestavros, "Popularity-aware GreedyDual-Size Web proxy caching algorithms," In Proceedings of ICDCS'2000.

....patterns. Such patterns not only lead to a better understanding of Web user behavior but also are valuable for various Web applications intended for characterizing the Web structure [11, 2] improving searching on the Web [10, 15, 17] and optimizing the Web performance via caching and prefetching [4, 12, 18, 21, 25]. The dramatic increase of HTTP tra#c on the Internet has resulted in wide spread use of large proxy servers as critical Internet infrastructure components. Web access logs collected at these proxy servers provide rich sources of information about Web user activities on the World Wide Web. As a ....

Shudong Jin and Azer Bestavros. Popularity-aware greedydual-size web proxy caching algorithms. In Proceedings of IEEE ICDS'00, April 2000.

....the write to cache operation available on some arrays (e.g. those from IBM [3] The difference is that we provide no way for the client to control which blocks the array should replace, and we trust the client to be well behaved. Recent studies of cooperative World Wide Web caching protocols [1, 20, 47] look at policies beyond LRU and MRU. Previously, analyses of web request traces [2, 5, 8] showed the file popularity distributions to be Zipf like [49] It is possible that schemes tuned for these workloads will perform as well for the sequential or random access patterns found in file system ....

S. Jin and A. Bestavros. Popularity-aware greedy-dual-size web proxy caching algorithms. In Proc. of the 20th Intl. Conf. on Distributed Computing Systems, pages 254--261. Apr. 2000.

....performance. The most common parameters take into consideration are frequency and recency. Replacement policies are designed to improve either HR or BHR or to obtain a compromise between both. Some replacement policies allow their users to define an arbitrary objective function for optimization [11, 6], but even in these cases their effectiveness is measured by HR and BHR. The optimization of both metrics is possible only if the cache space is enough to store all unique files. These caches are usually called infinite caches. There are many web caching replacement policies proposed in the ....

....even in these cases their effectiveness is measured by HR and BHR. The optimization of both metrics is possible only if the cache space is enough to store all unique files. These caches are usually called infinite caches. There are many web caching replacement policies proposed in the literature [1, 4, 3, 2, 18, 11, 6, 22, 16, 25, 26]. Both HR and BHR are considered the very important metrics for cache performance. The difference between BHR and HR for infinite caches, expressed in terms of the correlation between the document size and popularity, is present in [5] While their result is useful for analysing the properties of ....

[Article contains additional citation context not shown here]

S. Jin and A. Bestavros. Popularity-Aware GreedyDualSize Web Proxy Caching Algorithms. In Proceedings of ICDCS'

....bound of best achievable performance than a practical one. CHAPTER 2. WEB CACHING MODELING 24 In 1997, Cao and Irani proposed a group of web caching algorithms called GreedyDual Size algorithm [15] Since then, different enhanced versions of Greedy Dual Size algorithm have been proposed [16, 33]. The performance studies have shown that this group of algorithms is the champion of web caching algorithms [5, 33] 2.4.2 GD Size and GDSF In 1991, Young proposed the original Greedy Dual (GD) algorithm for caching in operating system [60] It is used in situations where pages in memory have a ....

.... and Irani proposed a group of web caching algorithms called GreedyDual Size algorithm [15] Since then, different enhanced versions of Greedy Dual Size algorithm have been proposed [16, 33] The performance studies have shown that this group of algorithms is the champion of web caching algorithms [5, 33]. 2.4.2 GD Size and GDSF In 1991, Young proposed the original Greedy Dual (GD) algorithm for caching in operating system [60] It is used in situations where pages in memory have a uniform size but have different costs for fetching them from the secondary storage. The algorithm assigns a key ....

S. Jin and A. Bestavros. Popularity-aware greedy-dual-size web proxy caching algorithms. In Proceedings of ICDCS'2000: The IEEE International Conference on Distributed Computing Systems, Taiwan, May 2000.

No context found.

S. Jin and A. Bestavros. Popularity-aware GreedyDual-size Web proxy caching algorithm. In Proceedings of ICDCS, April 2000.

No context found.

S. Jin and A. Bestavros. Popularity-aware GreedyDual-size Web proxy caching algorithm. In Proceedings of ICDCS, April 2000.

....By placing streaming media objects closer to customers, network bandwidth requirements are reduced, user perceived quality is improved, and demand on servers is decreased. Caching techniques have been widely explored for serving traditional Web objects such as HTML pages and image files [5, 6, 12, 24, 25]. For streaming media objects, caching becomes especially attractive due to their static nature, long duration and predictable sequential nature of access, and high network resource requirements. Paper Contributions and Overview: This paper proposes a novel technique that turns edge caches into ....

S. Jin and A. Bestavros. Popularity-aware GreedyDual-size Web proxy caching algorithm. In Proceedings of ICDCS, April 2000.

....mechanisms. By placing streaming media objects closer to customers, network bandwidth requirements are reduced, user perceived quality is improved, and demand on servers is decreased. Caching techniques have been widely explored for serving traditional Web objects such as HTML pages and image les [7, 9, 17, 34, 35]. For streaming media objects, caching becomes especially attractive due to the static nature of content, long duration and predictable sequential nature of accesses, and high network resource requirements. Ecient streaming media caching algorithms must consider several factors. They must ....

....techniques have been widely used for traditional Web content such as HTML pages and image les [1, 7, 10, 21] They reduce user perceived latency, network trac, and server load. Many studies have characterized HTTP requests [6, 8, 15] and have proposed cache management algorithms accordingly [9, 17, 34, 35]. HTTP request streams exhibit temporal locality of reference which leverages the e ectiveness of caching techniques. The most important property is the Zipf like [36] distribution of object popularity. With a Zipf like distribution, requests are concentrated on a few popular objects. Williams et ....

[Article contains additional citation context not shown here]

S. Jin and A. Bestavros. Popularity-aware GreedyDual-size Web proxy caching algorithm. In Proceedings of ICDCS, April 2000. 19

No context found.

S. Jin and A. Bestavros, "Popularity-Aware GreedyDual-Size Web Proxy Caching Algorithms," Proc. 20th Int'l Conf. Distributed Computing Systems (ICDCS '00), Apr. 2000.

No context found.

S. Jin and A. Bestavros, "Popularity-aware greedyDual-size Web proxy caching algorithms", Proceedings of 20th International Conference on Distributed Computing Systems, (ICDCS'2000.

No context found.

S. Jin and A. Bestavros, "Popularity-aware GreedyDual-Size Web proxy caching algorithms, " in Proceedings of IEEE ICDCS 2000.

No context found.

Shudong Jin and Azer Bestavros, " Popularity-Aware Greedy Dual-Size Web Proxy Caching Algorithms" , 20 International Conference on Distributed Computing Systems(ICDCS), 2000.

No context found.

S. Jin, A. Bestavros, Popularity-Aware Greedy Dual-Size Web Proxy Caching Algorithm, International Conference on Distributed Computing Systems, 2000.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC