Replica Placement in Data-Intensive Distributed Computing Systems

分布式计算系统是许多应用程序的首选平台。在这些系统中，应用程序是由竞争共享的异构资源（计算机，存储通信链路等）的大量用户提交的。因此，分布式系统可以被看作是由活跃用户共享的计算、存储和通信资源的集合。为此，需要开发新一代的算法和软件工具，以便通过将数据适当分配给相互竞争的应用程序和用户来有效利用这些系统。这个项目是朝着这个方向迈出的重要一步。

Distributed computing systems are the platform of choice for many applications. In these systems, applications are submitted by a large number of users that compete for the shared heterogeneous resources (computers, storage communication links, etc). Thus, a distributed system can be viewed as a collection of computing, storage and communication resources shared by active users. Towards this end, a new generation of algorithms and software tools need to be developed for the efficient utilisation of these systems through an appropriate allocation of data to competing applications and users. This project is a major step in this direction.