Efficient Scheduling of Resources for Parallel Query Processing on Grid-based Architecture
Khin Mar Soe, Aye Aye Nwe, Thandar Aung, Thinn Thu Naing, Ni Lar Thein · 2005
Parallel query processing is an increasingly attractive option for improving the performance of database systems. It can also be important in Grid since grid technologies have enabled sophisticated interaction and data sharing between resources that may belong to different departments or organizations. The decreasing cost of computing makes parallel query processing economically viable to reduce the response time of decision support queries by using parallel execution to exploit inexpensive resources. In this paper, the architecture of resource scheduling and site selection for parallel queries which are to be executed on the Grid is proposed. The main aims are to address the problems of choosing appropriate resources and matching sub plans with these resources. For this reason, decisions have to be made on allocating the available processors among a number of competing database operations running in parallel. For scheduling intra-query parallelism, the new approaches, resource-balancing based site selection algorithm which determines where to execute the upcoming operation sequence based on above architecture, and the resource assignment policy for handling the idle nodes’ efficiency well, are proposed. This paper also present the architecture of scheduling inter-query parallelism and proposed hierarchical model in order to fully exploit the available parallelism and interactions among the different queries.