T. Kelly and D. Reeves. Optimal web cache sizing: Scalable methods for exact solutions. Computer Communications, 24(2):163--173, 2001. 3.4 Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
On the Intrinsic Locality Properties of Web Reference.. - Fonseca, Almeida.. (2003) (3 citations) (Correct)
....and insights. For example, the nature of Web request streams has informed the development of cache replacement policies [1] 2] inter cache coordination protocols [3] and prefetching algorithms [4] It has even become possible to contemplate the cost optimal design of Web caches [5]. Nonetheless, the vast majority of the work so far on Web reference streams has dealt with individual streams in isolation, overlooking the fact that the Web is a system of clients, intermediaries, and servers, each of which may simultaneously emit or consume Web reference streams. In this work ....
.... and also because it can be directly used to generate synthetic workloads [15] Many studies have used stack distance to capture and characterize temporal locality in Web request streams [9] 13] 17] 18] and other authors have used the stack distance transformation as a tool for cache sizing [5] and workload generation. However, stack distance does not fit our needs because it cannot distinguish the causes of temporal locality directly. As already noted, the authors in [9] consider both popularity and correlation, and capture the difference between these two kinds of temporal locality ....
T. Kelly and D. Reeves, "Optimal Web cache sizing: Scalable methods for exact solutions," in Proceedings of the 5th Web Caching Workshop, May 2000.
Analyzing Web Robots and Their Impact on Caching - Almeida, Menascé.. (2001) (5 citations) (Correct)
....the number of objects it can hold. We consider all objects to be equally sized. This model, albeit simplistic, can give us insight on the behavior of the cache for the different streams. We use the marginal distribution of Least Recently Used (LRU) stack distances to determine the cache miss ratio [12, 14] under the LRU policy. If D is the random variable corresponding to the stack distances and FD is the cumulative distribution function of D, then the miss ratio m (x) for a cache of size x is given by P [D x] 1; FD (x) m (x) We start by noting the significant difference in the asymptotical miss ....
T. Kelly and D. Reeves. Optimal Web cache sizing: scalable methods for exact solutions. Computer Communications,Vol. 24, pages 163--173, February 2001.
Thin-Client Web Access Patterns: Measurements from a Cache-Busting .. - Kelly (2001) (7 citations) Self-citation (Kelly) (Correct)
....storage nor bandwidth are cheap or abundant. Therefore the relationship between browser cache size and performance is our primary concern. An efficient single pass simulation algorithm permits us to compute arbitrarily weighted hit rates at every cache size for each client in the WebTV trace [23]. It is straightforward to aggregate individual client success functions to obtain hit rate as a function of cache size for the entire client population, and this is the centerpiece of our analysis. Reference locality and document aliasing have performance implications, and we shall explore these ....
.... graphical evidence suggests that the distribution of LRU stack distances at the client may be lognormal (Figure 7, right; the data shown are from a sample of requests from a subset of BPS clients, described below) This distribution is closely related to the success function of an LRU cache [23,25] and is often used to measure temporal locality in reference streams [2, 9] Almeida et al. report that references reaching servers appear to have lognormal stack distance distributions, and that lognormal 0 0.25 0.5 0.75 1 1 10 2 10 4 10 6 10 8 10 10 P[X =x] Refs Bytes Figure 6: ....
[Article contains additional citation context not shown here]
T. Kelly and D. Reeves. Optimal Web cache sizing: Scalable methods for exact solutions. Computer Communications, 24:163--173, Feb. 2001.
Software Methods to Improve Data Locality and Cache Behavior - Beyls (2004) (Correct)
No context found.
T. Kelly and D. Reeves. Optimal web cache sizing: Scalable methods for exact solutions. Computer Communications, 24(2):163--173, 2001. 3.4
On the Optimization of Storage Capacity Allocation .. - Laoutaris.. (2003) (Correct)
No context found.
Kelly, T., Reeves, D.: Optimal web cache sizing: scalable methods for exact solutions. Computer Communications 24 (2001) 163--173
Joint Object Placement and Node Dimensioning for Internet.. - Laoutaris (2004) (Correct)
No context found.
T. Kelly and D. Reeves. Optimal web cache sizing: scalable methods for exact solutions. Computer Communications, 24(2):163--173, February 2001.
Software Methods to Improve Data Locality and Cache Behavior - Beyls (2004) (Correct)
No context found.
T. Kelly and D. Reeves. Optimal web cache sizing: Scalable methods for exact solutions. Computer Communications, 24(2):163--173, 2001. 3.4
Joint Object Placement and Node Dimensioning for Internet.. - Laoutaris (2004) (Correct)
No context found.
T. Kelly and D. Reeves. Optimal web cache sizing: scalable methods for exact solutions. Computer Communications, 24(2):163--173, February 2001.
Web Workloads Influencing Disconnected Service Access - Chandra (2001) (5 citations) (Correct)
No context found.
T. Kelly and D. Reeves. Optimal Web cache sizing: scalable methods for exact solutions. In Fifth International Web Caching and Content Delivery Workshop, 22-24 May 2000, Lisbon, Protugal. http://ai.eecs.umich.edu/~tpkelly/papers.
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