Storing and Replication in Topic-Based Publish/Subscribe Networks

Abstract

In current publish/subscribe networks messages are not stored and only active subscribers receive published messages. However, in a dynamic scenario a user may be interested in content published before the subscription time. In this paper, we introduce a mechanism that enables storing in such networks, while maintaining the main principle of loose-coupled and asynchronous communication. Furthermore, we propose a new storage placement and replication algorithm which differentiates classes of content and minimize the clients response latency. The performance of our proposed placement and replication algorithm and the proposed storing mechanism is evaluated via simulations and insights are given for future work.