Traditionally, Content Delivery Networks (CDNs) deploy proxy servers at strategic locations at the edge of the network to efficiently serve client requests. With the tremendous growth in multimedia applications and the number of clients accessing such applications, an edge proxy server itself may serve clients connected to it through a multi-hop network of heterogeneous links. Further, a special class of multimedia applications that can tolerate start up delays is emerging. In such applications, clients require a minimum acceptable quality (loss-free transmission at a minimum encoded rate ri) and the start of play back at a specific time (t + di) where t is the current time and di is the delay tolerance acceptable to client i. Our work deals with enhancing performance of such networks through a Hybrid Streaming Mechanism (HSM). In HSM a client’s request triggers the selection of an intermediate node as a streaming point to which multimedia contents are dynamically transferred from the proxy/source, and this streaming point streams the contents to the client. Transferred contents are temporarily cached at the streaming point to service future requests for the same content. HSM helps a Content Service Provider’s objective of satisfying as many client requests as possible and providing enhanced quality to clients given their delay tolerance. Simulation results demonstrate that by leveraging the delay tolerance of clients, and by combining the dynamic download and streaming mechanisms, HSM performs better than directly streaming from edge servers, serving on an average 40 % more client requests. 1.

Abstract—Geographically replicating popular objects in the Internet speeds up content distribution at the cost of keeping the replicas consistent and up-to-date. The overall effectiveness of replication can be measured by the total communication cost consisting of client accesses and consistency management, both of which depend on the locations of the replicas. This paper investigates the problem of placing replicas under the widely used TTL-based consistency scheme. A polynomial-time algorithm is proposed to compute the optimal placement of a given number of replicas in a network. The new replica placement scheme is compared, using real Internet topologies and Web traces, against two existing approaches which do not consider consistency management or assume invalidation-based consistency scheme. The factors affecting their performance are identified and discussed. Index Terms—Replication, replica placement, TTL-based consistency, data consistency. 1

This paper presents an algorithm to group correlated objects that are most likely to be requested by a client in a single session. Based on these groups, a centralized algorithm that determines the placements of objects to a cluster of web-servers is proposed to minimize latency. Due to the dynamic nature of the Internet traffic and the rapid changes in the access pattern of the World-Wide Web, we also propose a distributed algorithm where each site relies on some collected information to decide what object should be replicated at that site. The performance of the proposed algorithms is evaluated through a simulation study. Povzetek: Grupiranje objektov na spletu. 1

Web caching is an important technology for improving the scalability of Web services. One of the key problems in coordinated enroute Web caching is to compute the locations for storing copies of an object among the enroute caches so that some specified objectives are achieved. In this article, we address this problem for tree networks, and formulate it as a maximization problem. We consider this problem for both unconstrained and constrained cases. The constrained case includes constraints on the cost gain per node and on the number of object copies to be placed. We present dynamic programming-based solutions to this problem for different cases and theoretically show that the solutions are either optimal or convergent to optimal solutions. We derive efficient algorithms that produce these solutions. Based on our mathematical model, we also present a solution to coordinated enroute Web caching for autonomous systems as a natural extension of the solution for tree networks. We implement our algorithms and evaluate our model on different performance metrics through extensive simulation experiments. The implementation results show Preliminary versions of the results of this article have partially appeared in Li and Shen [2003a] on unconstrained enroute Web caching and in Li and Shen [2003b] on constrained enroute Web

(Internal Examiner)

Abstract not found

Abstract not found