Name
Abstract | ||
|---|---|---|
The ultimate goal of resource management in virtual computer and network systems is to reduce the amount of used resources to minimize the total cost of ownership (TCO). However, current methods to perform the adjusting of resource assignation are based solely on the usage level of resources themselves. They do not contemplate that there are other conditions that affect TCO, such as making changes in the system and the possibility of degrading the service (rejecting traffic). Those other conditions must be considered when deciding what resources must be adjusted and how in order to actually minimize the TCO. In this paper we propose a decision algorithm that considers, in addition to the cost of using resources, the additional costs of requesting the underlying provider to adjust the resources and the cost of degrading the service level of the system. The resource costs and service degradation are somewhat opposite, so avoiding resource costs will incur in service degradation and avoiding service degradation requires to increase resource costs. Thus, deciding the resource assignation is not a direct answer and depends on the running conditions of the controlled system. After each specific situation, our algorithm evaluates different options to adjust resources and chooses the one that results in the lower TCO. Therefore, our approach is evolutionary, as the system evolves gradually towards the optimum, even when considering the changing requirements of the running system. We demonstrate that a TCO reduction of more than 12 % can be achieved by using our algorithm in comparison to a traditional method that only considers resource utilization. |
Year | DOI | Venue |
|---|---|---|
2019 | 10.1109/ICIN.2019.8685844 | 2019 22nd Conference on Innovation in Clouds, Internet and Networks and Workshops (ICIN) |
Keywords | Field | DocType |
resource adjustment,resource management,virtual computer,network systems,TCO,resource assignation,resource costs,resource utilization,total cost-of-ownership minimization,decision algorithm,service level | Resource management,Service level,Virtual machine,Computer science,Total cost of ownership,Operations research | Conference |
ISSN | ISBN | Citations |
2162-3414 | 978-1-5386-8337-8 | 1 |
PageRank | References | Authors |
0.37 | 7 | 3 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Pedro Martinez-Julia | 1 | 109 | 20.06 |
| Ved P. Kafle | 2 | 209 | 35.01 |
| Hitoshi Asaeda | 3 | 21 | 6.31 |