Abstract: The cloud computing affords excessive prospects for the necessities of multifarious computing and consistent communication. Categorized by the increasing arrival rate of live content, the emergency applications that stance a great challenge how to publicize large-scale live content to interested users in a scalable and reliable manner. The publish/subscribe (pub/sub) model is broadly used for data dissemination of its capacity of impeccably escalating the system to colossal size. Most event matching services of existing pub/sub systems either lead to low matching throughput when matching a large number of twisted subscriptions when a large number of servers fail. SREM, a scalable and reliable event matching service for content-based pub/sub systems in cloud computing environment to realize low routing latency and consistent links between servers, a distributed overlay Skip Cloud to organize servers of SREM. Through a Partition technique, large-scale skewed subscriptions are charted into several subspaces, which safeguards high matching quantity and delivers multiple candidate servers for each event. Assessing the concert of SREM, 64 servers are organized and millions of live content items are tested in a Cloud Stack test bed. The tentative results shown that the traffic above of routing events in Skip Cloud is at least 60 percent smaller than in Chord overlay, the matching rate in SREM is at least 3.7 times and at most 40.4 times larger than the single-dimensional partitioning technique of Blue Dove based on various parameter settings. SREM facilitates the event loss rate to drop back to 0 in tens of seconds even if a large number of servers fail concurrently.
Keywords: Publish/subscribe, event matching, overlay construction, content space partitioning, cloud computing.