Browse by author
Lookup NU author(s): Ali Alssaiari, Dr Nigel Thomas
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0).
Power consumption in data centres receive a huge concern by data centre providers. This paper considers modelling a policy in PEPA to control powering on or off servers dynamically. The main focus was to identify and reflect over the trade-off between saving energy (by powering down servers) and performance cost. While powering down servers save energy, it could increase the performance cost. The experiment analyses the effect of the policy on energy consumption and performance cost. Different combination of dynamic and static servers used in this policy against different scenarios including changes job arrival rate, job arrival duration and the time that is needed by servers to fully power on and serve jobs. This gives an interesting outcome because every scenario is unique and therefore no server combinations does gives low energy and high performance in all scenario.
Author(s): Alssaiari A, Thomas N
Publication type: Conference Proceedings (inc. Abstract)
Publication status: Published
Conference Name: 32nd UK Performance Engineering Workshop
Year of Conference: 2016
Print publication date: 09/09/2016
Acceptance date: 12/08/2016
Date deposited: 31/10/2016
URL: http://computing.brad.ac.uk/ukpew2016/