Evaluation of a distributed numerical simulation optimization approach applied to aquifer remediation

In this paper we evaluate a distributed approach which uses numerical simulation and optimization techniques to automatically find remediation solutions to a hypothetical contaminated aquifer. The repeated execution of the numerical simulation model of the aquifer through the optimization cycles tends to be computationally expensive. To overcome this drawback, the numerical simulations are executed in parallel using a network of heterogeneous workstations. Performance metrics for heterogeneous environments are not trivial; a new way of calculating speedup and efficiency for Bag-of-Tasks (BoT) applications is proposed. The performance of the parallel approach is evaluated.