Master Thesis MSTR-2018-56

BibliographyMaa, Marius: In-network complex event processing using advanced data-plane programming.
University of Stuttgart, Faculty of Computer Science, Electrical Engineering, and Information Technology, Master Thesis No. 56 (2018).
87 pages, english.

Event detection latency is an important aspect of complex event processing which could be improved significantly by utilizing the capabilities of networking devices, which already exist in computer networks. For this reason, this thesis presents a new language for expressing complex event detection rules along with P4CEP, a compiler which translates these rules into P4 code, which can be used for integrating complex event detection into the data-plane. P4CEP reads a CEP rule definition, builds a finite state machine for event detection and then converts the rules into P4 code and forwarding table entries. P4CEP was tested and evaluated with the Netronome SmartNIC and compared against the state-of-the-art CEP framework Apache Flink. In a simple test case and with the Netronome SmartNIC, an event detection latency of 18 s was measured for P4CEP compared to 232 s with Apache Flink, which is a 93% decrease in latency.

Full text and
other links
Department(s)University of Stuttgart, Institute of Parallel and Distributed Systems, Distributed Systems
Superviser(s)Rothermel, Prof. Kurt; Kohler, Thomas; Mayer, Ruben
Entry dateJune 4, 2019
   Publ. Computer Science