Abstract
One of the main advantages brought by the Internet of Things (IoT) is the possibility of having large amounts of data from several sources that allow us, once analyzed, to make decisions in various domains in real time. This implies the need to be able to process large volumes of data in more or less limited processing times depending on the application domain. In this sense, complex event processing (CEP), used in conjunction with an enterprise service bus (ESB), has proven to be very efficient in multiple domains. In search for greater efficiency, some CEP engines offer the option of using flow-based programming (FBP) rather than their traditional programming using CEP together with an event bus. However, its use, while it may be more efficient, can lead to other limitations. In this article, we analyze and describe the performance and limitations of using a CEP engine with an ESB versus a CEP engine with FBP. This will allow developers to decide which option is more convenient for their IoT system depending on the application domain and its specific needs.
Highlights
Large amounts of data are generated from multiple sources that are waiting to be processed in order to obtain greater knowledge of the domain in question and to make profitable decisions
If we can implement the system in RQ1 without major limitations, can we improve the performance of a Complex Event Processing (CEP) application integrated with an Enterprise Service Bus (ESB) by using Flow-Based Programming (FBP), avoiding having to use the ESB; and, if so, how much does it improve our systems performance?
In response to the research questions posed in the Introduction, we can assert that, as anticipated, a system based on the CEP-Dataflow implementation can be used to promptly detect situations of interest from incoming events in streaming in real time
Summary
Large amounts of data are generated from multiple sources that are waiting to be processed in order to obtain greater knowledge of the domain in question and to make profitable decisions. In order to answer these research questions, the main aims of this paper are, on the one hand, to implement two equivalent systems where one uses a CEP with an ESB and the other a CEP with dataflows and FBP, as well as to define a benchmark that allows us to evaluate their performance. We will analyze the difficulties and limitations in the implementation of both systems and their performance according to the benchmark in question to evaluate in which scenarios it is more convenient to use either For this purpose, we have chosen to use the Esper CEP engine [17]: Esper is a highly scalable and open source Java-based software engine for CEP that can rapidly process and analyze large volumes of incoming IoT data in real time.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
