A Markovian sensibility analysis for parallel processing scheduling on GNU/Linux

  • Authors:

  • Regiane Y. Kawasaki;Luiz Affonso Guedes;Diego L. Cardoso;Carlos R. L. Francês;Glaucio H. S. Carvalho;Solon V. Carvalho;João C. W. A. Costa;Marcelino S. Silva

  • Affiliations:

  • Department of Electrical and Computing Engineering, Federal University of Pará (UFPA), Belém, PA, Brazil;Department of Computing Engineering and Automation, Federal University of Rio Grande do Norte (UFRN), Natal, RN, Brazil;Department of Electrical and Computing Engineering, Federal University of Pará (UFPA), Belém, PA, Brazil;Department of Electrical and Computing Engineering, Federal University of Pará (UFPA), Belém, PA, Brazil;Department of Electrical and Computing Engineering, Federal University of Pará (UFPA), Belém, PA, Brazil;Computing and Applied Mathematics Laboratory (LAC), National Institute for Space Research (INPE), São José dos Campos, SP, Brazil;Department of Electrical and Computing Engineering, Federal University of Pará (UFPA), Belém, PA, Brazil;Department of Electrical and Computing Engineering, Federal University of Pará (UFPA), Belém, PA, Brazil

  • Venue:
  • ISPA'06 Proceedings of the 2006 international conference on Frontiers of High Performance Computing and Networking
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

Parallel Computing has become a powerful tool to overcome certain types of computational problems in many areas such as engineering, especially due to the increasing diversity of platforms for execution of this type of application. The use of parallel computing over LANs and WANs is an alternative in the universe of dedicated environments (parallel machines and clusters), but, in some cases, it needs to imply QoS (Quality of Service) parameters, so it can execute efficiently. In this scenario, the deployment of resource allocation scheme plays an important role in order to satisfy the QoS requirements for parallel applications. In this paper we propose and present Markovian models for resource allocation (CPU allocation) schemes in a GPOS (General Purpose Operating Systems), aiming at offering an optimization method which makes the efficient performance of parallel and interactive applications feasible.