Power efficient scheduling heuristics for energy conservation in computational grids
| Autor: | Hesham El-Rewini, Abdul Aziz |
|---|---|
| Rok vydání: | 2011 |
| Předmět: |
Computer science
business.industry Distributed computing Cloud computing Computational resource computer.software_genre Grid Supercomputer Failover Fair-share scheduling Theoretical Computer Science Scheduling (computing) Energy conservation Grid computing Hardware and Architecture Redundancy (engineering) business Heuristics computer Software Information Systems |
| Zdroj: | The Journal of Supercomputing. 57:65-80 |
| ISSN: | 1573-0484 0920-8542 |
| DOI: | 10.1007/s11227-011-0559-7 |
| Popis: | Computation grids and computational clouds are becoming increasingly popular in the organizations which require massive computational capabilities. Building such infrastructures makes a lucrative business case, thanks to availability of cheap hardware components and affordable software. Maintaining computational grids or cloud, however, require careful planning as in such dedicated environments, round-the-clock availability of workstations is very crucial. Ensuring uninterrupted availability, not only demands mechanism for failover redundancy but also results in constant power drainage. The tradeoff between the cost and the performance is the constant dilemma that the operations of the data centers face today. In this paper, we propose various heuristics for power-aware scheduling algorithms for scheduling jobs with dependent tasks onto the computational grid. We formulate the problem as a multi-objective function which results in various cost-performance tradeoffs each lying within the solution boundary. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|