Scheduling in Distributed Computing Systems

We study scheduling in large-scale systems such as clusters, clouds and datacenters. Much of our research centers around the design, implementation, and analysis of the KOALA multischeduler that has been deployed in the Dutch DAS multicluster system. 

  • Resource management and scheduling of data-processing frameworks

An example of our recent work is the design and analysis of dynamic resource balancing of cloud resources among MapReduce frameworks:

Bogdan Ghit, Nezih Yigitbasi, Alexandru Iosup, and Dick Epema, "Balanced Resource Allocations across Multiple Dynamic MapReduce Clusters," ACM Sigmetrics 2014.

  •  Portfolio scheduling 

A single (adaptive) scheduling policy is not always enough to cover all circumstances in which a system might operate. We work on a technique called portfolio scheduling for having systems change once in a while radically from one policy to another.

Kefeng Deng, Junqiang Song, and Kaijun Ren, and Alexandru Iosup, Exploring Portfolio Scheduling for Long-Term Execution of Scientific Workloads in IaaS Clouds, ACM/IEEE Conference on High Performance Computing (SC13).

Selected Publications on Scheduling

Please note: This page contains links to PostScript files of articles that may be covered by copyright. You may browse the articles at your convenience. (In the same spirit as you may read a journal or a proceeding article in a public library). Retrieving, copying, or distributing these files, however, may violate the copyright protection law. We recommend that the user abides international law in accessing this article list.

  • Alexey Ilyushkin, Ahmed Ali-Eldin, Nikolas Herbst, Alessandro Papadopoulos, Bogdan Ghit, Dick Epema, Alexandru Iosup (2017). An Experimental Performance Evaluation of Autoscaling Policies for Complex Workflows. In Proceedings of the 8th ACM/SPEC on International Conference on Performance Engineering, L'Aquila, Italy. [ Bibtex ]
Tribler logo
Bit-measure logo
Koala Grid Scheduler logo
Grench Mark logo
DAS 5 logo