An efficient energy saving task consolidation algorithm for cloud computing systems

Task consolidation is a process of maximizing resource utilization in a cloud system. However, maximum usage of resources does not necessarily imply that there will be proper use of energy as some resources which are sitting idle, also consume considerable amount of energy. Recent studies show that...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:2014 International Conference on Parallel, Distributed and Grid Computing s. 262 - 267
Hlavní autoři: Panda, Sanjaya K., Jana, Prasanta K.
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 01.12.2014
Témata:
ISBN:1479976822, 9781479976829
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Task consolidation is a process of maximizing resource utilization in a cloud system. However, maximum usage of resources does not necessarily imply that there will be proper use of energy as some resources which are sitting idle, also consume considerable amount of energy. Recent studies show that energy consumption due to idle resources is approximately 1 to 20%. So, the idle resources are assigned with some tasks to utilize the idle period, which in turn reduces the overall energy consumption of the resources. Note that higher resource utilization merely leads to high energy consumption. So, the tasks are likely to be assigned to all the resources for the proper use of energy. In this paper, we propose an energy saving task consolidation (ESTC) which minimizes the energy consumption by utilizing the idle period of the resources in a cloud environment. ESTC achieves it by assigning few tasks to all available resources to overcome the idleness of the resources. In addition to this, it calculates the energy consumption on arrival of a task to make the scheduling assessment. We perform extensive experiments to measure the performance of ESTC and we compare it with the recent energy-aware task consolidation (ETC) algorithm. The results show that the proposed algorithm outperforms ETC in terms of energy consumption and the total number of task completion.
ISBN:1479976822
9781479976829
DOI:10.1109/PDGC.2014.7030753