Abstract
Grid resources are non-dedicated, and thus grid users are forced to compete with resource owners for idle CPU cycles. As a result, the turnaround times of both the grid jobs and the owners' jobs are invariably delayed. To resolve this problem, the current study proposes a progressive multi-layer resource reconfiguration framework, designated as PMR-G, in which intra- and inter-site reconfiguration strategies are employed to adapt grid users' jobs dynamically to changes in the available CPU resources at each node. The experimental results show that PMR-G enables the idle CPU cycles of a resource to be fully exploited by grid users with minimum interference to the resource owner's jobs.
| Original language | English |
|---|---|
| Pages (from-to) | 662-673 |
| Number of pages | 12 |
| Journal | Future Generation Computer Systems |
| Volume | 25 |
| Issue number | 6 |
| DOIs | |
| Publication status | Published - 2009 Jun 1 |
Fingerprint
All Science Journal Classification (ASJC) codes
- Software
- Hardware and Architecture
- Computer Networks and Communications
Cite this
}
A progressive multi-layer resource reconfiguration framework for time-shared grid systems. / Chen, Po Cheng; Chang, Jyh Biau; Liang, Tyng Yeu; Shieh, Ce-Kuen.
In: Future Generation Computer Systems, Vol. 25, No. 6, 01.06.2009, p. 662-673.Research output: Contribution to journal › Article
TY - JOUR
T1 - A progressive multi-layer resource reconfiguration framework for time-shared grid systems
AU - Chen, Po Cheng
AU - Chang, Jyh Biau
AU - Liang, Tyng Yeu
AU - Shieh, Ce-Kuen
PY - 2009/6/1
Y1 - 2009/6/1
N2 - Grid resources are non-dedicated, and thus grid users are forced to compete with resource owners for idle CPU cycles. As a result, the turnaround times of both the grid jobs and the owners' jobs are invariably delayed. To resolve this problem, the current study proposes a progressive multi-layer resource reconfiguration framework, designated as PMR-G, in which intra- and inter-site reconfiguration strategies are employed to adapt grid users' jobs dynamically to changes in the available CPU resources at each node. The experimental results show that PMR-G enables the idle CPU cycles of a resource to be fully exploited by grid users with minimum interference to the resource owner's jobs.
AB - Grid resources are non-dedicated, and thus grid users are forced to compete with resource owners for idle CPU cycles. As a result, the turnaround times of both the grid jobs and the owners' jobs are invariably delayed. To resolve this problem, the current study proposes a progressive multi-layer resource reconfiguration framework, designated as PMR-G, in which intra- and inter-site reconfiguration strategies are employed to adapt grid users' jobs dynamically to changes in the available CPU resources at each node. The experimental results show that PMR-G enables the idle CPU cycles of a resource to be fully exploited by grid users with minimum interference to the resource owner's jobs.
UR - http://www.scopus.com/inward/record.url?scp=63649130426&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=63649130426&partnerID=8YFLogxK
U2 - 10.1016/j.future.2009.01.002
DO - 10.1016/j.future.2009.01.002
M3 - Article
VL - 25
SP - 662
EP - 673
JO - Future Generation Computer Systems
JF - Future Generation Computer Systems
SN - 0167-739X
IS - 6
ER -