A swarm-inspired data center consolidation methodology
Cristina Bianca Pop, Ionut Manuel Anghel, Tudor Cioara, Ioan Salomie, Iulia Vartic · 2012
This paper proposes a swarm-inspired data center consolidation methodology which aims at reducing the power consumption in data centers while ensuring the workload execution within the pre-established performance parameters. Each data center server is managed by an intelligent agent that deals with its power efficiency by implementing a bird's migration-inspired behavior to decide on the appropriate server consolidation actions. The selected actions are executed to achieve an optimal utilization of server computing resources thus lowering power consumption. The data center servers self-organize in logical clusters according to the birds V-formation self-organizing migration model. The results are promising showing that the swarm-inspired data center consolidation methodology optimizes the utilization ratio of the data center computing resources and achieves estimated power savings of about 16%.