Système de messagerie événementiel (publish/subscribe)

This is a multi-faceted project on unifying event dissemination and event processing in a single, high performance, platform. Students will be in charge of investigating popular applications, formulating novel problems, designing, implementing, and evaluating their solution.

The objectives of this project include the following. Note that students are expected to choose a subset of goals to fulfill:

1- Integrate event processing operations (e.g., aggregation, top-k filtering, stream joins) with event dissemination (e.g., publish/subscribe). The integrated solution should provide better performance than a disconnected baseline.

2- Integrate event dissemination with emerging networking technologies, such as Softwaredefined Networking (SDN), Network Function Virtualization (NFV), and Future Internet Architectures (e.g. CCNx). Demonstrate the performance benefits of leveraging advanced networking.

3- Support techniques for machine learning (e.g., convex optimization) with on-the-fly processing in an event dissemination system. Demonstrate the increase in velocity in the unified system.

Connaissances requises

Degree in computer science, software engineering
Experience in system research, deployment, evaluation
Strong programming skills in C++, Java, or Python
Familiarity with Big data systems (Storm, Spark, Hadoop), SDN (OpenFlow, OpenVSwitch), or machine learning systems (TensorFlow, Torch, ...)

Programme d'études visé

Maîtrise avec mémoire, Doctorat

Domaines de recherche

Technologies de l'information et des communications

Financement

Bourses disponibles pour maîtrises et doctorats 

Autres informations

Début : Automne 2020  

https://fuseelab.github.io/index_fr.html

Personne à contacter

Kaiwen Zhang | kaiwen.zhang@etsmtl.ca