Matchmaking with limited knowledge of resources on clouds and grids

Clouds and grids are distributed resource infrastructures that are subjected to both on demand as well as Advance reservation (AR) requests. Matchmaking or assigning an appropriate resource to an advance reservation request is an important operation performed by the resource management middleware. Most existing research on matchmaking strategies uses a priori knowledge of the local scheduling policy used at a resource in the cloud/grid. However, exact knowledge of the scheduling policy used at every resource is unlikely to be available on real heterogeneous systems deploying a variety of different resources. This paper presents a matchmaking strategy that does not use detailed knowledge of the scheduling policies used at resources. Results of performance simulations for the strategy are very encouraging.