Grand challenge: real-time soccer analytics leveraging low-latency complex event processing

Authors:
Martin Jergler;Christoph Doblander;Mohammedreza Najafi;Hans-Arno Jacobsen
Affiliations:
Technical University Munich, Munich, Germany;Technical University Munich, Munich, Germany;Technical University Munich, Munich, Germany;Technical University Munich, Munich, Germany
Venue:
Proceedings of the 7th ACM international conference on Distributed event-based systems
Year:
2013

Citing 4
Cited 0

Programming Erlang: Software for a Concurrent World

Programming Erlang: Software for a Concurrent World
Kilim: Isolation-Typed Actors for Java

ECOOP '08 Proceedings of the 22nd European conference on Object-Oriented Programming
Multi-query Stream Processing on FPGAs

ICDE '12 Proceedings of the 2012 IEEE 28th International Conference on Data Engineering
The DEBS 2013 grand challenge

Proceedings of the 7th ACM international conference on Distributed event-based systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, we present a real-time capable event-based system, which is tailored towards analytical query processing in the context of soccer games. The main challenge is to meet the application's strict real-time and low-latency requirements in face of streams of high-velocity sensor data. We describe a workflow-like architecture for query processing based on a publish/subscribe model. Queries are structured into computational tasks that are arranged sequentially and/or in parallel. Tasks are connected by preallocated ring buffers providing total event ordering and fast as well as decoupled event access. Our evaluation results show the effectiveness of the proposed system in terms of low-latency processing under real-time conditions. Speeding up the system by a factor of 50 compared to real-time introduces almost no latency overhead.