Abstract:
Complex Event Processing (CEP) is one of the most
rapidly emerging fields in data processing. Processing of high
volume of events to derive higher level events is a vital part of
several applications. The use-cases found in Business applications,
financial trading applications, operational analytics applications
and business activity monitoring applications are directly related
to complex event processing.
This paper discusses different design decisions associated with
CEP Engines, and proposes to improve CEP performance by
using a stream-processing-style pipelines. Furthermore, the paper
discusses Siddhi, a CEP Engine that implements pipelines for
stream processing and presents a performance study that exhibits
that Siddhi CEP Engine has significantly improved performance.
Primary contributions of this paper are, a critical analysis
of the CEP and Event Stream Processing engine architecture
and identifying areas for improvements, implementing those
improvements through Siddhi, and demonstrating the soundness
of those suggestions through empirical evidence.