Complex events processing in data streaming

Prof. Assaf Schuster | Computer Science


Information and Computer Science

The Technology

Complex event processing (CEP) algorithms are widely employed in applications where arbitrarily complex patterns of data items must be promptly and efficiently detected in massive data streams. CEP algorithms treat data items as events, as new events are detected they are combined into higher-level complex events matching the user-specified patterns. Modern CEP engines are required to efficient tracking thousands of patterns simultaneous in multiple high-speed input streams of events. Most CEP systems also operate under tight real-time constraints, where a pattern match must be reported within milliseconds of its occurrence. In many real-life scenarios the data characteristics are never known in advance or are subject to frequent on-the-fly changes.

The technology introduces novel algorithms and method that enable reprocessing on the data streams themselves reducing dramatically off-line analysis, thus making it efficient and faster.


  • Utilizing these improved plans allows CEP systems to achieve a performance boost of up to several orders of magnitude in terms of throughput, latency and memory consumption
  • Significantly improves the quality of the generated pattern detection plans by considering a broader solution space
  • Optimization algorithm can operate under arbitrarily tight time constraints due to the inherent balance between optimization time and solution quality

Applications and Opportunities

  • Any application that requires monitoring and tracking events of interest in massive data streams
  • Capital market
  • Information security
  • Sensors network
  • Credit card fraud detection
  • Inventory management (such as retail chains)
  • Autonomous cars
  • Medical information systems
  • Smart cities
  • Wearable products
  • Internet of Things (Industry 4.0)
arrow Business Development Contacts
Motti Koren
Director of Business Development, Life Sciences