Scheduling Challenges in a Shared Private Cloud Infrastructure

Warning

This publication doesn't include Faculty of Arts. It includes Institute of Computer Science. Official publication website can be found on muni.cz.

Authors	KLUSÁČEK Dalibor PARÁK Boris HEJTMÁNEK Lukáš
Year of publication	2017
Type	Article in Periodical
Magazine / Source	CGW Workshop'17 Proceedings
MU Faculty or unit	Institute of Computer Science
Citation
Keywords	cloud; grid; resource sharing; utilization; fairness; scheduling; OpenNebula
Description	This work describes several scheduling challenges that appear in a shared computing system that hosts different applications (grid jobs vs. cloud VMs) with different demands and usage patterns. Using the real-life data from the system, we identify several problems (such as unfairness and low CPU utilization) and propose solutions to alleviate their negative impacts.
Related projects:	CERIT Scientific Cloud