%0 Journal Article
%T Fast Optimal Replica Placement with Exhaustive Search Using Dynamically Reconfigurable Processor
%A Hidetoshi Takeshita
%A Sho Shimizu
%A Hiroyuki Ishikawa
%A Akifumi Watanabe
%A Yutaka Arakawa
%A Naoaki Yamanaka
%A Kosuke Shiba
%J Journal of Computer Networks and Communications
%D 2011
%I Hindawi Publishing Corporation
%R 10.1155/2011/707592
%X This paper proposes a new replica placement algorithm that expands the exhaustive search limit with reasonable calculation time. It combines a new type of parallel data-flow processor with an architecture tuned for fast calculation. The replica placement problem is to find a replica-server set satisfying service constraints in a content delivery network (CDN). It is derived from the set cover problem which is known to be NP-hard. It is impractical to use exhaustive search to obtain optimal replica placement in large-scale networks, because calculation time increases with the number of combinations. To reduce calculation time, heuristic algorithms have been proposed, but it is known that no heuristic algorithm is assured of finding the optimal solution. The proposed algorithm suits parallel processing and pipeline execution and is implemented on DAPDNA-2, a dynamically reconfigurable processor. Experiments show that the proposed algorithm expands the exhaustive search limit by the factor of 18.8 compared to the conventional algorithm search limit running on a Neumann-type processor. 1. Introduction Content delivery networks (CDNs) [1¨C3] are being developed to improve the userĄ¯s experience when downloading voluminous files such as music and videos, a rapidly growing component of the traffic on the Internet. A CDN consists of two types of servers: origin server and replica server. The original data is stored in the origin server and then copied to the replica servers, which are geographically distributed. A user requesting content is connected to a replica server automatically selected by the network, which then sends the content to the user. Replica selection is based on the distance between the server and the user, and usually, the closest server is selected [4]. One important issue in CDN performance is replica placement [5]. The problem is in deciding which servers are to hold which replicas. Replica servers cache the original serversĄ¯ contents to prevent traffic congestion and to maintain user performance. They allow CDN providers to minimize the capital expenditures (CapEx) and operational expenditures (OpEx). Note that cache size is restricted; no replica server can hold all of the contents held by the origin server. For each content, we must pick those servers, not all, that will hold the replica. In addition, in order to achieve adequate user performance, each replica server must have a limited delivery area. The delivery area is expressed as the distance from the replica server. Each user must lie within the delivery area of at least one replica
%U http://www.hindawi.com/journals/jcnc/2011/707592/