GDL

Efficient Resource Matching in Heterogeneous Grid Using Resource Vector

ISSN:0975-3826
Journal from gdlhub / 2017-08-14 11:52:33
Oleh : Srirangam V Addepallil, Per Andersenand George L Barnes, International Journal of Computer Science & Information Technology
Dibuat : 2012-06-26, dengan 1 file

Keyword : SGE, LSF, Venus, ENDyne, Condor.
Subjek : Efficient Resource Matching in Heterogeneous Grid Using Resource Vector
Url : http://airccse.org/journal/jcsit/0203csit1.pdf
Sumber pengambilan dokumen : Internet

In this paper, a method for efficient scheduling to obtain optimum job throughput in a distributed campus


grid environment is presented; Traditional job schedulers determine job scheduling using user and job


resource attributes. User attributes are related to current usage, historical usage, user priority and


project access. Job resource attributes mainly comprise of soft requirements (compilers, libraries) and


hard requirements like memory, storage and interconnect. A job scheduler dispatches jobs to a resource


if a job’s hard and soft requirements are met by a resource. In current scenario during execution of a job,


if a resource becomes unavailable, schedulers are presented with limited options, namely re-queuing job


or migrating job to a different resource. Both options are expensive in terms of data and compute time.


These situations can be avoided, if the often ignored factor, availability time of a resource in a grid


environment is considered. We propose resource rank approach, in which jobs are dispatched to a


resource which has the highest rank among all resources that match the job’s requirement. The results


show that our approach can increase throughput of many serial / monolithic jobs.

Deskripsi Alternatif :

In this paper, a method for efficient scheduling to obtain optimum job throughput in a distributed campus


grid environment is presented; Traditional job schedulers determine job scheduling using user and job


resource attributes. User attributes are related to current usage, historical usage, user priority and


project access. Job resource attributes mainly comprise of soft requirements (compilers, libraries) and


hard requirements like memory, storage and interconnect. A job scheduler dispatches jobs to a resource


if a job’s hard and soft requirements are met by a resource. In current scenario during execution of a job,


if a resource becomes unavailable, schedulers are presented with limited options, namely re-queuing job


or migrating job to a different resource. Both options are expensive in terms of data and compute time.


These situations can be avoided, if the often ignored factor, availability time of a resource in a grid


environment is considered. We propose resource rank approach, in which jobs are dispatched to a


resource which has the highest rank among all resources that match the job’s requirement. The results


show that our approach can increase throughput of many serial / monolithic jobs.

Beri Komentar ?#(0) | Bookmark

Properti	Nilai Properti
ID Publisher	gdlhub
Organisasi	International Journal of Computer Science & Information Technology
Nama Kontak	Herti Yani, S.Kom
Alamat	Jln. Jenderal Sudirman
Kota	Jambi
Daerah	Jambi
Negara	Indonesia
Telepon	0741-35095
Fax	0741-35093
E-mail Administrator	elibrarystikom@gmail.com
E-mail CKO	elibrarystikom@gmail.com