download0 view127
twitter facebook

공공누리This item is licensed Korea Open Government License

Title
Exploiting resource profiling mechanism for large-scale scientific computing on grids
Author(s)
MD AZAM HOSSAIN김직수NGUYEN NGOC CAO황순욱
Publication Year
2016-07-09
Abstract
Large-scale scientific applications from various scientific domains (e.g., astronomy, physics, pharmaceuticals, chemistry, etc.) usually require substantial amounts of computing resources and storage space. International Grid computing resources can be a viable choice for supporting these challenging applications so that effectively locating suitable computing resources with minimal allocation overhead can be crucial. However, Grid resource availability is highly unstable and current Grid Information Service (GIS) cannot provide accurate state information. This can make it very difficult for users to schedule the jobs on the Grid system and to map tasks on appropriate available resources. In this paper, we present SCOUT system that can periodically profile Grid computing elements based on available number of CPU cores and average response time, and monitor the performance of each CE in the Virtual Organizations (VO). Micro-benchmark experimental results demonstrate that leveraging profiled data by SCOUT can improve the success rate of task executions and reduce the average response time.
Keyword
Resource profiling; Job scheduling; Grid computing; Scout; Scientific computing
Journal Title
Cluster Computing: The Journal of Networks, Software Tools and Applications
ISSN
1386-7857
Files in This Item:
There are no files associated with this item.
Appears in Collections:
7. KISTI 연구성과 > 학술지 발표논문
URI
https://repository.kisti.re.kr/handle/10580/14516
http://www.ndsl.kr/ndsl/search/detail/article/articleSearchResultDetail.do?cn=NART76443557
Export
RIS (EndNote)
XLS (Excel)
XML

Browse