This paper introduces a framework for evaluating replica placement algorithms (RPA) for content delivery networks (CDN) as well as RPAs from other fields that might be applicable to current or future CDNs. First, the framework classifies and qualitatively compares RPAs using a generic set of primitives that capture problem definitions and heuristics. Second, it provides estimates for the decision times of RPAs using an analytic model. To achieve accuracy, the model takes into account disk accesses and message sizes, in addition to computational complexity and message numbers that have been considered traditionally. Third, it uses the "goodness" of produced placements to compare RPAs even when they have different problem definitions. Based on these evaluations, we identify open issues and potential areas for future research.

The rapid growth of time-critical information services and business-oriented applications is making quality of service (QoS) support increasingly important in content distribution. This paper investigates the problem of placing object replicas (e.g., web pages and images) to meet the QoS requirements of clients with the objective of minimizing the replication cost. We consider two classes of service models: replica-aware service and replica-blind service. In the replica-aware model, the servers are aware of the locations of replicas and can therefore direct requests to the nearest replica. We show that the QoS-aware placement problem for replica-aware services is NP-complete. Several heuristic algorithms for efficient computation of suboptimal solutions are proposed and experimentally evaluated. In the replica-blind model, the servers are not aware of the locations of replicas or even their existence. As a result, each replica only serves the requests flowing through it under some given routing strategy. We show that there exist polynomial optimal solutions to the QoS-aware placement problem for replicablind services. Efficient algorithms are proposed to compute the optimal locations of replicas under different cost models.

A commonly employed abstraction for studying the object placement problem for the purpose of Internet content distribution is that of a distributed replication group. In this work the initial model of distributed replication group of Le#, Wolf, and Yu (IEEE TPDS '93) is extended to the case that individual nodes act selfishly, i.e., cater to the optimization of their individual local utilities. Our main contribution is the derivation of equilibrium object placement strategies that: (a) can guarantee improved local utilities for all nodes concurrently as compared to the corresponding local utilities under greedy local object placement; (b) do not su#er from potential mistreatment problems, inherent to centralized strategies that aim at optimizing the social utility; (c) do not require the existence of complete information at all nodes. We develop a baseline computationally e#cient algorithm for obtaining the aforementioned equilibrium strategies and then extend it to improve its performance with respect to fairness. Both algorithms are realizable in practice through a distributed protocol that requires only limited exchange of information.

Abstract—The rapid growth of new information services and business-oriented applications entails the consideration of quality of service (QoS) in content distribution. This paper investigates the QoS-aware replica placement problems for responsiveness QoS requirements. We consider two classes of service models: replica-aware services and replica-blind services. In replica-aware services, the servers are aware of the locations of replicas and can therefore optimize request routing to improve responsiveness. We show that the QoS-aware placement problem for replica-aware services is NP-complete. Several heuristic algorithms for fast computation of good solutions are proposed and experimentally evaluated. In replica-blind services, the servers are not aware of the locations of replicas or even their existence. As a result, each replica only serves the requests flowing through it under some given routing strategy. We show that there exist polynomial optimal solutions to the QoS-aware placement problem for replica-blind services. Efficient algorithms are proposed to compute the optimal locations of replicas under different cost models. Index Terms—Content distribution, replication, placement, quality of service, dynamic programming, NP-complete. 1

Abstract — Network coding provides elegant solutions to many data transmission problems. The usage of coding for distributed data storage has also been explored. In this work, we study a joint storage and transmission problem, where a source transmits a file to storage nodes whenever the file is updated, and clients read the file by retrieving data from the storage nodes. The cost includes the transmission cost for file update and file read, as well as the storage cost. We show that such a problem can be transformed into a pure flow problem and is solvable in polynomial time using linear programming. Coding is often necessary for obtaining the optimal solution with the minimum cost. However, we prove that for networks of generalized tree structures, where adjacent nodes can have asymmetric links between them, file splitting — instead of coding — is sufficient for achieving optimality. In particular, if there is no constraint on the numbers of bits that can be stored in storage nodes, there exists an optimal solution that always transmits and stores the file as a whole. The proof is accompanied by an algorithm that optimally assigns file segments to storage nodes. I.

The placement of data to maximize the performance and minimize the cost of a computing system is an optimization problem that has been studied extensively in several fields, including distributed databases, storage systems and, more recently, content delivery networks. However, little has been done to compare the various approaches and their applicability to different systems.

A file storage scheme is proposed for networks containing heterogeneous clients. In the scheme, the performance measured by file-retrieval delays degrades gracefully under increasingly serious faulty circumstances. The scheme combines coding with storage for better performance. The problem is NP-hard for general networks; and this paper focuses on tree networks with asymmetric edges between adjacent nodes. A polynomial-time memory-allocation algorithm is presented, which determines how much data to store on each node, with the objective of minimizing the total amount of data stored in the network. Then a polynomial-time data-interleaving algorithm is used to determine which data to store on each node for satisfying the quality-of-service requirements in the scheme. By combining the memory-allocation algorithm with the data-interleaving algorithm, an optimal solution to realize the file storage scheme in tree networks is established.

Abstract—In this paper, we discuss and compare several policies to place replicas in tree networks, subject to server capacity and Quality-of-Service (QoS) constraints. The client requests are known beforehand, while the number and location of the servers are to be determined. The standard approach in the literature is to enforce that all requests of a client be served by the closest server in the tree. We introduce and study two new policies. In the first policy, all requests from a given client are still processed by the same server, but this server can be located anywhere in the path from the client to the root. In the second policy, the requests of a given client can be processed by multiple servers. One major contribution of this paper is to assess the impact of these new policies on the total replication cost. Another important goal is to assess the impact of server heterogeneity, both from a theoretical and a practical perspective. In this paper, we establish several new complexity results and provide several efficient polynomial heuristics for NP-complete instances of the problem. These heuristics are compared one to the other, and their absolute performance is assessed by comparison with the optimal solution provided by an integer linear program.

Most important commercial Web sites maintain multiple replicas of their server infrastructure to increase both reliability and performance. In this paper, we study how many replicas should be used and where they should be placed in order to improve client network performance, including both the latency (e.g., round-trip time) between clients and the replicas, and the bandwidth performance between them. This study is based on a large scale measurement study from an 18-node infrastructure, which reveals for the first time the distribution of today's Internet end-user access bandwidth. For example, we find that 50% of end users have access bandwidth less than 4.2Mbps. Using a greedy algorithm, we show that the first five replicas dominate latency optimization in our measurement infrastructure, while the first two replicas dominate bandwidth optimization. We also found that geographic diversity does not help as much for bandwidth optimization as it does for latency. To determine the proper trade-off between latency and bandwidth, we use a simplified TCP model to show that, when content size is less than 10KB, the deployment should focus on optimizing latency, while for content sizes larger than 1MB, the deployment should focus on optimizing bandwidth.

Abstract — This paper studies the QoS-aware replica placement problem. Although there has been much work on replica placement problem, most of them concerns average system performance and ignores quality assurance issue. Quality assurance is very important, especially in heterogeneous environments. We propose a new heuristic algorithm that determines the positions of replicas in order to satisfy the quality requirements imposed by data requests. The experimental results indicate that the proposed algorithm finds a near-optimal solution effectively and efficiently for algorithm can also adapt to various parallel and distributed environments. I.