JAMILA: a usable batch job management system to coordinate heterogeneous clusters and diverse applications over grid or cloud infrastructure

  • Authors:
  • Juan Peng;Xiaoyi Lu;Boqun Cheng;Li Zha

  • Affiliations:
  • Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China;Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China;Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China;Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China

  • Venue:
  • NPC'10 Proceedings of the 2010 IFIP international conference on Network and parallel computing
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

Usability is an important feature of Grids or Clouds to end users, who may not be computer professionals but need to use massive machines to compute their jobs. For meeting various computing or management requirements, heterogeneous clusters with diverse Distributed Resource Management Systems (D-RMS) and applications are needed to supply computing services in Grids or Clouds. The heterogeneity of clusters and diversity of applications are easy to cause Grid or Cloud systems hard to use. This paper presents a usable batch job management system, called JAMILA, to coordinate heterogeneous clusters and diverse applications over Grid or Cloud infrastructure. JAMILA aims to implement a uniform job management mechanism to integrate different kinds of DRMSes and it is designed to supply non-professional users from various fields with a professional but usable high performance computing environment. Some key technologies of JAMILA have been used in China National Grid and an experimental Cloud-oriented infrastructure.