ScholarLens Back to search
2014•Unpublished venueRequires access

A PGAS Execution Model for Efficient Stencil Computation on Many-Core Processors

Mitsuru Ikei, Mitsuhisa Sato

Open publisher page 0 citations

Abstract

A efficient PGAS execution model on many-core processor for stencil computation is proposed and implemented. We use XcalableMP as a base language and we modify its runtime well fit in many-core processors. The runtime uses processes for parallel execution and global arrays of the stencil codes are broken into blocked sub-arrays placed on shared memory. Using two stencil codes, Laplace and Himeno, we evaluated its performance. In the evaluation, we show (1) Blocking improves locality of memory access during computation therefore improves total CPU execution time. (2) Direct data access using shared memory can relieve communication burden of sub-array halo exchanges.

About this research paper

What this paper is about

A efficient PGAS execution model on many-core processor for stencil computation is proposed and implemented. We use XcalableMP as a base language and we modify its runtime well fit in many-core processors. The runtime uses processes for parallel execution and global arrays of the stencil codes are broken into blocked sub-arrays placed on shared memory. Using two stencil codes, Laplace and Himeno, we evaluated its performance. In the evaluation, we show (1) Blocking improves locality of memory access during computation therefore improves total CPU execution time. (2) Direct data access using shared memory can relieve communication burden of sub-array halo exchanges.

Why it matters

A significance statement is not available in the OpenAlex record.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

A efficient PGAS execution model on many-core processor for stencil computation is proposed and implemented. We use XcalableMP as a base language and we modify its runtime well fit in many-core processors. The runtime uses processes for parallel execution and global arrays of the stencil codes are broken into blocked sub-arrays placed on shared memory. Using two stencil codes, Laplace and Himeno, we evaluated its performance. In the evaluation, we show (1) Blocking improves locality of memory access during computation therefore improves total CPU execution time. (2) Direct data access using shared memory can relieve communication burden of sub-array halo exchanges.

Key concepts: Stencil, Computer science, Parallel computing, Partitioned global address space, Computation, Shared memory, Locality, Multi-core processor

Related papers

Related paper 1Related paper 2
Back to paper searchBrowse research topicsOriginal source
A PGAS Execution Model for Efficient Stencil Computation on Many-Core Processors — Research Paper | ScholarLens