Affiliations: [a] Facultad de Informática, Universidad Politécnica de Madrid, Madrid, Spain | [b] Dpto. De Electrónica y Computadores, Universidad de Cantabria, Santander, Spain
Correspondence:
[*]
Corresponding author: Pilar Herrero, Facultad de Informática, Universidad Politécnica de Madrid, Campus de Montegancedo s/n, Boadilla del Monte, Madrid 28660, Spain. Tel.: +34 91 336 74 56; Fax: +34 91 336 65 95; E-mail: pherrero@fi.upm.es
Abstract: Cooperation among computational nodes to solve a common parallel application is one of the most outstanding features of grid environments. However, from the performance point of view, it is very important that this cooperation will be made in an equilibrated way, because otherwise some nodes can be overloaded whereas other nodes can be underused. This paper presents CAMBLE (Cooperative Awareness Model for Balancing the Load in grid Environments), which applies some theoretical principles of awareness models to promote an efficient, autonomous, equilibrate and cooperative task delivery in grid environments. This cooperative task management has been implemented and tested in a real and heterogeneous grid infrastructure, composed of several VO, with very successful results. This paper presents some of these outcomes while emphasizes on the overhead and efficacy of the system using this model.
Keywords: Grid computing, awareness of interaction, computer supportive cooperative work
