2nd International ICST Conference on Autonomic Computing and Communication Systems

Research Article

Measuring the Quality of an Artificial Hormone System based Task Mapping

Download532 downloads
  • @INPROCEEDINGS{10.4108/ICST.AUTONOMICS2008.4662,
        author={Uwe Brinkschulte and Alexander von Renteln and Mathias Pacher},
        title={Measuring the Quality of an Artificial Hormone System based Task Mapping},
        proceedings={2nd International ICST Conference on Autonomic Computing and Communication Systems},
        publisher={ICST},
        proceedings_a={AUTONOMICS},
        year={2010},
        month={5},
        keywords={artificial hormone system real-time task mapping decentralized control loops},
        doi={10.4108/ICST.AUTONOMICS2008.4662}
    }
    
  • Uwe Brinkschulte
    Alexander von Renteln
    Mathias Pacher
    Year: 2010
    Measuring the Quality of an Artificial Hormone System based Task Mapping
    AUTONOMICS
    ICST
    DOI: 10.4108/ICST.AUTONOMICS2008.4662
Uwe Brinkschulte1,*, Alexander von Renteln1,*, Mathias Pacher1,*
  • 1: Institute for Process Control, Automation, and Robotics, University of Karlsruhe (TH), Germany
*Contact email: brinks@ira.uka.de, renteln@ira.uka.de, pacher@ira.uka.de

Abstract

The idea of Organic Computing is a trend to counter the problems arising from the fact that computing systems are getting smaller and smaller, and we will soon be surrounded by large numbers of little computers which will be hard to configure, maintain, and control. We reintroduce an organic middleware - the artificial hormone system (AHS) which can map tasks onto a grid of heterogeneous processing elements while providing the system with self-X properties and even guaranteeing upper bounds for the self-configuration and self-healing. This paper investigates the quality of task mappings on a grid of heterogeneous processing elements. An algorithm is proposed to measure the quality of such task mappings. Experiments with randomly generated configurations will show results of mappings done by our artificial hormone system and compare them with ordinary load balancing.