Event-Based Information Fusion for the Self-Adaptive Cloud

Project: Dissertation

Research areas and keywords

UKÄ subject classification

  • Computer Systems

Description

Successful self-adaptive resource provisioning in the cloud relies on accurate tracking of workload variations and timely detection of changes in the infrastructure. The general estimation problem is very challenging due to the massive number of observable events in various subsystems, each containing some useful information. In this project, we will develop novel, event-based estimation techniques for information fusion in cloud server systems. Our starting point will be the family of Monte Carlo-based inference methods known as Particle Filters, which will be adapted to handle event-based measurements from different sources and with different time scales. The results will enable more responsive and exact decision making in the autonomous cloud.

During 2020 we have summarized our study on how particle filtering techniques can be adapted to better handle event-based measurements. In a parallel line of work, we have developed a testbed for experimenting with scalable cloud applications. We have also started to model and identify networks of microservices using queueing models and measurements from Cloud applications.

Layman's description

En välfungerande självinställande resursallokering i molnet kräver noggrann följning av lastvariationer och snabb upptäckt av förändringar i infrastrukturen. Det generella skattningsproblemet är mycket utmanande eftersom det genereras ett enormt antal potentiellt intressanta händelser per tidsenhet i olika delsystem. I detta projekt kommer vi att utveckla nya, händelsebaserade skattningsmetoder för informationsfusion i molnsystem. Vår utgångspunkt är partikelfilter (Monte Carlo-baserade inferensmetoder), som kommer att anpassas till händelsebaserade mätningar från olika källor och med olika tidsskalor. Resultaten kommer att möjliggöra mer alert och mer exakt resursfördelning i det autonoma molnet.
StatusActive
Effective start/end date2017/09/012022/09/01

Participants

Related research output

Tommi Nylander, Johan Ruuskanen, Karl-Erik Årzén & Martina Maggio, 2020, Proceedings of the 2020 ACM/SPEC International Conference on Performance Engineering. p. 24-35

Research output: Chapter in Book/Report/Conference proceedingPaper in conference proceeding

Johan Ruuskanen & Anton Cervin, 2020, 2020 19th European Control Conference (ECC). 8 p.

Research output: Chapter in Book/Report/Conference proceedingPaper in conference proceeding

Tommi Nylander, Johan Ruuskanen, Karl-Erik Årzén & Martina Maggio, 2020, ACM/SPEC International Conference on Performance Engineering Companion (ICPE ’20 Companion). 3 p.

Research output: Chapter in Book/Report/Conference proceedingPaper in conference proceeding

View all (7)