A Scalable and Reliable Matching Service for Content-Based Publish Subscribe Systems - 2015
Characterized by the increasing arrival rate of live content, the emergency applications create a great challenge: the way to disseminate giant-scale live content to interested users in a very scalable and reliable manner. The publish/subscribe (pub/sub) model is widely used for data dissemination as a result of of its capability of seamlessly expanding the system to huge size. However, most event matching services of existing pub/sub systems either cause low matching throughput when matching a large variety of skewed subscriptions, or interrupt dissemination when a giant number of servers fail. The cloud computing provides great opportunities for the requirements of complex computing and reliable communication. In this paper, we tend to propose SREM, a scalable and reliable event matching service for content-primarily based pub/sub systems in cloud computing setting. To realize low routing latency and reliable links among servers, we tend to propose a distributed overlay SkipCloud to organize servers of SREM. Through a hybrid area partitioning technique HPartition, large-scale skewed subscriptions 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 are extensively studied. To evaluate the performance of SREM, 64 servers are deployed and many live content items are tested during a CloudStack testbed. Underneath various parameter settings, the experimental results demonstrate that the traffic overhead of routing events in SkipCloud is at least sixty percent smaller than in Chord overlay, the matching rate in SREM is at least 3.seven times and at most 40.4 times larger than the only-dimensional partitioning technique of BlueDove. Besides, SREM enables the event loss rate to drop back to zero in tens of seconds even if a large variety of servers fail simultaneously.
Did you like this research project?
To get this research project Guidelines, Training and Code... Click Here