Resumen de Data flow parallel evaluation of Datalog Queries

José Francisco Aldana Montes , José María Troya Linero

• A Dataflow model to obtain parallelism in the evaluation of Datalog is presented. This model performs query evaluation as a data flow through a network of communicating concurrent processes capable of solving the query. Thus it is possible to increase the intra-query parallelism available for the query evaluation. This process network is based on the Intensional Data Base definition, plus the concrete query to be evaluated. An algorithm to optimize the evaluation is described which is based on process network rewriting. This utilizes information in the query bindings to be evaluated in order to eliminate the non-generating part of the process network and to filter data flow via the same. A naive cost model to cope with the load balancing problem in the specific context is described. A load balancig algorithm is presented and discussed.