World-Wide Web proxy servers that cache documents can potentially reduce three quantities: the number of requests that reach popular servers, the volume of network traffic resulting from document requests, and the latency that an end-user experiences in retrieving a document. This paper examines the first two using the measures of cache hit rate (or fraction of client-requested URLs returned by the proxy) and weighted cache hit rate (or fraction of client-requested bytes returned by the proxy). A client request for an uncached document may cause the removal of one or more cached documents. Variable document sizes and types allow a rich variety of policies to select a document for removal, in contrast to policies for CPU caches or demand paging, that manage homogeneous objects. We present a taxonomy of removal policies. Through trace-driven simulation, we determine the maximum possible hit rate and weighted hit rate that a cache could ever achieve, and the removal policy that maximizes hi...

Abstract—With the recent explosion in usage of the World Wide Web, the problem of caching Web objects has gained considerable importance. Caching on the Web differs from traditional caching in several ways. The nonhomogeneity of the object sizes is probably the most important such difference. In this paper, we give an overview of caching policies designed specifically for Web objects and provide a new algorithm of our own. This new algorithm can be regarded as a generalization of the standard LRU algorithm. We examine the performance of this and other Web caching algorithms via event- and trace-driven simulation.

Given the continued growth of the World-Wide Web, performance of Web servers is becoming increasingly important. File caching can be used to reduce the time that it takes a Web server to respond to client requests, by storing the most popular files in the main memory of the Web server, and by reducing the volume of data that must be transferred between secondary storage and the Web server. In this paper, we use trace-driven simulation to evaluate the effects of various replacement, threshold, and partitioning policies on the performance of a Web server. The workload traces for the simulations come from Web server access logs, from six different Internet Web servers. The traces represent three different orders of magnitude in server activity and two different orders of magnitude in time duration. The results from our simulation study show that frequency-based caching strategies, using a variation of the LFU (Least Frequently Used) replacement policy, perform the best for the Web server ...

Abstract not found

Understanding WWW traffic characteristics is key to improving the performance and scalability of the Web. In the first part of this thesis, Web proxy workloads from different levels of a caching hierarchy are used to understand how the workload characteristics change across different levels of a caching hierarchy. The main observations of this study are: HTML and image documents account for 95% of the documents seen in the workload; the distribution of transfer sizes of documents is heavy-tailed, with the tails becoming heavier as one moves from the client side to the server side of the network; the popularity profile of documents does not precisely follow the Zipf distribution; one-timers account for approximately 70% of the documents referenced; concentration of references is less at proxy caches than at servers, and concentration of references is higher at lower-level proxies than at higher-level proxies; there appears to be no correlation between document modification rate and document pop...

Web caching has been considered as a powerful solution to deal with the growth of web traffic. Several studies have shown that caching documents throughout the Internet can save network bandwidth and reduce document access latency [9, 8, 10]. However, this technique has introduced new problems such as maintaining the document coherency and selecting the next document to be removed. With the continuous increase in demand for documents, the web cache servers are becoming the new bottleneck. A need for better resource management is becoming urgent in order to reduce the overhead sustained by web cache servers. In this paper, a number of web replacement policies are discussed and compared on the basis of trace-driven simulations. The impact of the web cache server configuration is pointed out through a set of experiments that use the cache size as a tuning parameter.

The rapid growth of the Web has sparked significant research activity on improving the performance and scalability of the Web. Much of this work focuses on Web caching: client-side caching, server-side caching, proxy-level caching, and multilevel caching hierarchies. Single-level caching and multi-level caching hierarchies combined with well-informed cache management strategies are obvious keys to the performance and scalability of the Web. The first part of this thesis research describes the design and use of a synthetic Web proxy workload generator called ProWGen to evaluate the effectiveness of Web proxy caching architectures. The synthetic workload generator can be used to closely match the salient characteristics of empirical Web proxy workloads (e.g., one-timers, Zipf-like referencing behaviour, heavy-tailed file size distribution, temporal locality), or to generate multiple workloads that differ from each other in carefully controlled ways (e.g., Zipf slope, tail index, correlation ...

Abstract. ThesuccessoftheWorldWideWebmeasuredintermsofthe number of its users and of the resulting traffic increase is only commensurate to the patience required when sitting in front of one’s computer, waiting for a document to be down-loaded. If one could identify the typical access patterns for a set of documents on a Web server, the server could use or extend the existing protocols to accordingly pre-fetch or push documents to the browsers and proxy servers. In this paper, we present and evaluate a strategy for making Web servers “pushier”. Which document is to be pushed is determined by a set of association rules mined from a sample of the access log of the Web server. Once a rule of the form “Document A → Document B ” has been identified and selected, the Web server decides to push “Document2 ” if “Document1” is requested. The strategy is individual user oriented while not ignoring the aggregate perspective. We evaluate the effectiveness and the cost of such a strategy for two architectures: a two tier “Web server / Web browser ” architecture, and a three tier “Web server / proxy server / Web browser ” architecture. We consider different settings in the architectures as well as refinements of the strategy taking into account the size of the documents. 1

We analyze the arrival rate of accesses to Web proxy caching servers. The results show that the data display strong periodic autocorrelation. The examined data sets show a consistent behavior in terms of having periods corresponding to daily and weekly cycles that can be explained in terms of daily and weekly cyclic behavior of Web users. While these results confirm the correlation in the network traffic noticed by other researchers, we emphasize that this correlation also is periodic. A new approach is introduced to model data that exhibit such characteristics by a combination of Fourier and statistical analysis techniques. The source of high correlation in the data is shown to come from the periodic (deterministic) part. Synthesized data that results from this modeling approach is shown to have a long-range dependent and self-similar behavior.

Storage management is important to the performance of DBMS. This paper gives a comprehensive overview of storage management in relational DBMS. The storage management is divided into three levels (logical, physical in-memory, and physical on-disk) and discussed separately. The logical level caches logical units (tuples or index values) of the DBMS based on the logical information to improve buffer pool usage efficiency. The physical in-memory level caches physical pages directly in the buffer pool to reduce disk accesses. Many algorithms have been designed for the general buffer or for the buffer pool of DBMS. Complex algorithms can achieve better performance than simple LRU or CLOCK algorithms. However, the overhead is high and the advantage diminishes when the buffer is big. The physical on-disk level borrows many techniques from file systems and storage systems. Because of the excellent potential write performance of log-structured organization, the Log-structured File System and related topics are discussed. Each level has its advantage and limitation for further improving performance. To achieve better performance of storage management systems of DBMS, all these three levels should be considered. 1