Semantically partitioned peer to peer complex event processing

Abstract

Scaling Complex Event Processing applications is inherently problematic. Many state of the art techniques for scaling use filtering on producers, vertical scaling, or stratification of an Event Processing Network. The solutions usually aren't distributed and require centralized coordination. In this paper, we are introducing a technique for scaling Complex Event Processing in a distributed fashion and by taking semantic information of events into account.We are introducing two CEP models for scaling CEP architectures, providing core algorithms, and evaluating their performance. © Springer International Publishing Switzerland 2014.