Search for: All records

Multiobjective optimization problems (MOPs) are common across many science and engineering fields. A multiobjective optimization algorithm (MOA) seeks to provide an approximation to the tradeoff surface between multiple, possibly conflicting, objectives. Many MOPs are the result of objective functions that require the evaluation of a computationally expensive numerical simulation. Solving these large and complex problems requires efficient coordination between the MOA and the computationally expensive cost functions. In this work, a recently proposed MOA is integrated into the libEnsemble software library, which coordinates extreme scale resources for large ensemble computations. Efficient integration requires fundamental changes to the underlying MOA. The convergence and performance results for the integrated and original MOA are compared on a set of benchmark problems. 

Variability in the execution time of computing tasks can cause load imbalance in high-performance computing (HPC) systems. When configuring system- and application-level parameters, engineers traditionally seek configurations that will maximize the mean computational throughput. In an HPC setting, however, high-throughput configurations that do not account for performance variability could result in poor load balancing. In order to determine the effects of performance variance on computationally expensive numerical simulations, the High-Performance LINPACK solver is optimized by using multiobjective optimization to maximize the mean and minimize the standard deviation of the computational throughput on the High-Performance LINPACK benchmark. We show that specific configurations of the solver can be used to control for variability at a small sacrifice in mean throughput. We also identify configurations that result in a relatively high mean throughput, but also result in a high throughput variability.