Conference Paper

Scheduling in Bag-of-Task Grids: The PAUÁ Case.

Univ. Fed. de Campina Grande, Brazil;
DOI: 10.1109/SBAC-PAD.2004.37 Conference: 16th Symposium on Computer Architecture and High Performance Computing (SBAC-PAD 2004), 27-29 October 2004, Foz do Iguacu, Brazil
Source: DBLP

ABSTRACT In this paper we discuss the difficulties involved in the scheduling of applications on computational grids. We highlight two main sources of difficulties: 1) the size of the grid rules out the possibility of using a centralized scheduler; 2) since resources are managed by different parties, the scheduler must consider several different policies. Thus, we argue that scheduling applications on a grid require the orchestration of several schedulers, with possibly conflicting goals. We discuss how we have addressed this issue in the context of PAUA, a grid for Bag-of-Tasks applications (i.e. parallel applications whose tasks are independent) that we are currently deploying throughout Brazil.

1 Bookmark
 · 
99 Views
  • [Show abstract] [Hide abstract]
    ABSTRACT: A 'Grid' is an infrastructure for resource sharing. It is used for large-scale data processing, many of the applications being scientific ones. The processing of a job on a conventional network takes place on a single computer, whereas the processing of a job on a Grid may be split between several computers. A Grid is, therefore, capable of running applications that are too demanding for a stand-alone computer or a conventional network of computers. The resources making up a Grid need to be managed to provide a good quality of service. Availability is complicated by the fact that a single application may request multiple resource types from multiple locations. Grid resource allocation is regarded as a vital component of a Grid infrastructure. Many methods have been developed for Grid resource allocation. In this paper, we propose a new Grid resource allocation method (Modified Least Cost Method) that assigns tasks to computing nodes in a way that is close to the optimum. The proposed method is an advanced form of the Least Cost Method. In this paper we evaluate Modified Least Cost Method for Grid resource allocation scenario. This paper also proposes a Linear Programming (LP) model for Grid scheduling and includes a comparative performance analysis of our proposed Resource Allocation method with the existing ones.
    Cyber-Enabled Distributed Computing and Knowledge Discovery (CyberC), 2010 International Conference on; 01/2010
  • Source
    Escola Regional de Informática de Minas Gerais; 01/2005

Full-text (2 Sources)

Download
78 Downloads
Available from
May 30, 2014