Resource Brokering in Grid Environments using Evolutionary Algorithms

The subject of this paper is the resource management for a grid system that is primarily intended to support computationally expensive tasks like simulations and optimisations on a grid. Applications are represented as workflows that can be decomposed into single grid jobs. The main task of the resource management is resource brokering to optimise a global schedule for all requesting grid jobs and all requested resources. The present contribution shall illustrate the necessity of planning and optimising resource allocation in a grid. Requirements to be met by a resource management system will be defined. These requirements are comparable with the requirements on planning systems in other fields, e.g. production planning systems. Here, various methods have already been developed for optimised planning. Suitable methods are Evolutionary Algorithms (EA). Consequently, a global optimising resource broker (GORBA) is proposed that deploys EA. The paper focuses on the concept of GORBA and the implementation of GORBA prototype.

[1]  Wolfgang Süß,et al.  A Grid Environment for Simulation and Optimization and a First Implementation of a Biomedical Application , 2004, OTM Workshops.

[2]  Rajkumar Buyya,et al.  Economic-based Distributed Resource Management and Scheduling for Grid Computing , 2002, ArXiv.

[3]  Wilfried Jakob.  Eine neue Methodik zur Erhöhung der Leistungsfähigkeit Evolutionärer Algorithmen durch die Integration lokaler Suchverfahren , 2004 .

[4]  Rajkumar Buyya,et al.  Nature's heuristics for scheduling jobs on Computational Grids , 2000 .

[5]  Christian Blume.  GLEAM - A System for Simulated 'Intuitive Learning' , 1990, PPSN.

[6]  C. Blume,et al.  Deutliche Senkung der Produktionskosten durch Optimierung des Ressourceneinsatzes , 1994 .

[7]  Christian Blume,et al.  GLEAM - An Evolutionary Algorithm for Planning and Control Based on Evolution Strategy , 2002, GECCO Late Breaking Papers.

[8]  Achim Streit,et al.  Scheduling in HPC Resource Management Systems: Queuing vs. Planning , 2003, JSSPP.

[9]  H. H. Rosenbrock,et al.  An Automatic Method for Finding the Greatest or Least Value of a Function , 1960, Comput. J..

[10]  Hans-Paul Schwefel,et al.  Evolution and Optimum Seeking: The Sixth Generation , 1993 .

[11]  Ian Foster,et al.  A quality of service architecture that combines resource reservation and application adaptation , 2000, 2000 Eighth International Workshop on Quality of Service. IWQoS 2000 (Cat. No.00EX400).

[12]  Ramin Yahyapour,et al.  Design and evaluation of job scheduling strategies for grid computing , 2000, GRID.

[13]  Wilfried Jakob.  HyGLEAM - An Approach to Generally Applicable Hybridization of Evolutionary Algorithms , 2002, PPSN.