No abstract

This paper examines the initial parallel implementation of SCATTER, a computationally intensive inelastic neutron scattering routine with polycrystalline averaging capability, for the General Utility Lattice Program (GULP). Of particular importance to structural investigation on the atomic scale, this work identifies the computational features of SCATTER relevant to a parallel implementation and presents initial results from performance tests on multi-core and multi-node environments. Our initial approach exhibits near-linear scalability up to 256 MPI processes for a significant model.

Widespread heterogeneous parallelism is unavoidable given the emergence of General-Purpose computing on graphics processing units (GPGPU). The characteristics of a Graphics Processing Unit (GPU)—including significant memory transfer latency and complex performance characteristics—demand new approaches to ensuring that all available computational resources are efficiently utilised. This paper considers the simple case of a divisible workload based on widely-used numerical linear algebra routines and the challenges that prevent efficient use of all resources available to a naive SPMD application using the GPU as an accelerator. We suggest a possible queue monitoring strategy that facilitates resource usage with a view to balancing the CPU/GPU utilisation for applications that fit the pipeline parallel architectural pattern on heterogeneous multicore/multi-node CPU and GPU systems. We propose a stochastic allocation technique that may serve as a foundation for heuristic approaches to balancing CPU/GPU workloads.

This paper explores the early implementation of highperformance routines for the solution of multiple large Hermitian eigenvector and eigenvalue systems on a Graphics Processing Unit (GPU). We report a performance increase of up to two orders of magnitude over the original EISPACK routines with a NVIDIA Tesla C2050 GPU, potentially allowing an order of magnitude increase in the complexity or resolution of a neutron scattering modeling application.

No abstract