AN ADAPTABLE AND PREDICTABLE MATCHING SERVICE FORCONTENTBASED PUBLISHING DATA SYSTEMS

Abstract

Characterized by the increasing arrival rate of live content, the emergency applications pose a great challenge: how to disseminate large-scale live content to interested users in a scalable and reliable manner. Thepublish/subscribe (pub/sub) model is widely used for data dissemination because of its capacity of seamlesslyexpanding the system to massive size. However, most event matching services of existing pub/sub systems either lead to low matching throughput when matching a large number of skewed subscriptions, or interrupt dissemination when a large number of servers fail. The cloud computing provides great opportunities for therequirements of complex computing and reliable communication. In this paper, we propose SREM, a scalableand reliable event matching service for content-based pub/sub systems in cloud computing environment. Toachieve low routing latency and reliable links among servers, we propose a distributed overlay Skip Cloudtoorganize servers of SREM. Through a hybrid space partitioning technique HPartition, large-scale skewedsubscriptions are mapped into multiple subspaces, which ensures high matching throughput and provides multiple candidate servers for each event. Moreover, a series of dynamics maintenance mechanisms areextensively studied. To evaluate the performance of SREM, 64 servers are deployed and millions of livecontent items are tested in a Cloud Stack test bed.