SC14 New Orleans, LA

The International Conference for High Performance Computing, Networking, Storage and Analysis

XcalableACC – a Directive-Based Language Extension for Accelerated Parallel Computing.

Authors: Hitoshi Murai (RIKEN), Masahiro Nakao (RIKEN), Takenori Shimosaka (RIKEN), Akihiro Tabuchi (University of Tsukuba), Taisuke Boku (University of Tsukuba), Mitsuhisa Sato (University of Tsukuba)

Abstract: Accelerated parallel computers (APC) such as GPU clusters are emerging as an HPC platform. This paper proposes XcalableACC, a directive-based language extension to program APCs. XcalableACC is a combination of two existing directive-based languages, XcalableMP and OpenACC, and has two additional functions: data/work mapping among multiple accelerators and direct communication between accelerators. The result of the preliminary evaluation shows that XcalableACC is a promising means to program APCs. In our poster, the basic concept of XcalableACC and the result of preliminary evaluation will be presented.

Poster: pdf
Two-page extended abstract: pdf

Poster Index