Direkt zum InhaltDirekt zur SucheDirekt zur Navigation
▼ Zielgruppen ▼

Humboldt-Universität zu Berlin - Mathematisch-Naturwissenschaftliche Fakultät - Process-Driven Architectures

Event Stream Processing (SE)

Prof. Dr. Matthias Weidlich

 

Content

Sensing of data is a major trend these days. According to ABI Research, by 2020, around 40 billion wireless devices are assumed to be connected to the Internet and many of them are expected to continuously emit events. Data stream processing and complex event processing (here jointly referred to as event stream processing) are technologies that help to make sense of these events, by filtering the data, transforming events, and matching event query patterns against a set of incoming event streams. Systems for event stream processing have been successfully used in various domains: news agencies exploit real-time data from social networks; urban transportation is guided by events emitted by street-level sensors; data from real-time location systems (RTLS) is used to optimise the operation of hospitals; and information on financial transactions is leveraged to identify credit card fraud. Yet, the increasing volume, velocity, variety and distribution of event sources imposes challenges for the design and implementation of event stream processing systems. To cope with these requirements, various competing approaches have been proposed in the literature, each taking particular design decisions, for instance, in terms of the time model (causal vs. absolute vs. interval), the deployment (clustered vs. centralised), the event model (discrete vs. probabilistic) and the interaction with event sources (push vs. pull). In this seminar, basic aspects of event stream processing will be explored by means of recent scientific papers.

 

Topics

Introductory slides incl. available topics

Notes on reviews

Please submit (email to matthias.weidlich@hu) your ranked selection of 3 topics by Monday, April 25, 2016.

 

 

Topic Assignment and Timeline

12.05.2016 

  • Topic 2 (Abadi 2003): Thomas Schlegel
  • Topic 3 (Tucker 2003): Dimitar Dimitrov 
 
19.05.2016
  • Topic 4 (Arasu 2006): Frank Lange
  • Topic 5 (Wu 2006): Simon Pizonka 
 
26.05.2016
  • Topic 6 (Mei 2009): Robin Ellerkmann
  • Topic 8 (Dindar 2013): Marc Kewitz 
 
02.06.2016: No meeting 
 
09.06.2016
  • Topic 12 (Liu 2009): Martin Fobian (cancelled)
  • Remarks on reviewing
 
16.06.2016: TAG DER INFORMATIK
23.06.2016: No meeting
30.06.2016: No meeting
 
07.07.2016:
  • Topic 14 (Arasu 2004): Arne Hoffmann
  • Topic 15 (Li 2015): Felix Fischer
  • Topic 12 (Liu 2009): Martin Fobian

 

14.07.2016

  • Topic 17 (Tangwongsan 2015): Nils Goldammer
  • Topic 21 (Tatbul 2003): Matthias Hoeschel
 
 

Organisation

Dates Thu 15-17
Location RUD 26, 1.303
AGNES 3313083
Degree
Informatik, Master of Science (M.Sc.); Informatik, Master of Education (M.Ed.); Wirtschaftsinformatik, Master of Science (M.Sc.). The related area of specialisation is "Daten- und Wissensmanagement".
Credit Points 5 LP (2 SWS)