An adaptive offline implementation selector for heterogeneous parallel platforms. (November 2018)

• Record Type: Journal Article
• Authors: del Rio Astorga, David; Dolz, Manuel F; Sánchez, Luis Miguel; Fernández, Javier; García, J Daniel
• Other Names: Balaji Pavan guest-editor.; Leung Kai-Cheung guest-editor.; Huang Zhiyi guest-editor.; García-Blas Javier guest-editor.; Brown Christopher guest-editor.
• Is Part Of: International journal of high performance computing applications. Volume 32:Number 6(2018)
• Page Start: 854
• Page End: 863
• Publication Date: 2018-11
• Subjects: Implementation selector -- heterogeneous platforms -- auto-tuning -- C++ attributes -- C++ concepts; High performance computing -- Periodicals; Supercomputers -- Periodicals; 004.1105
• DOI: 10.1177/1094342017698746
• Languages: English
• ISSNs: 1094-3420

Abstract

Heterogeneous parallel platforms, comprising multiple processing units and architectures, have become a cornerstone in improving the overall performance and energy efficiency of scientific and engineering applications. Nevertheless, taking full advantage of their resources comes along with a variety of difficulties: developers require technical expertise in using different parallel programming frameworks and previous knowledge about the algorithms used underneath by the application. To alleviate this burden, we present an adaptive offline implementation selector that allows users to better exploit resources provided by heterogeneous platforms. Specifically, this framework selects, at compile time, the tuple device-implementation that delivers the best performance on a given platform. The user interface of the framework leverages two C++ language features: attributes and concepts. To evaluate the benefits of this framework, we analyse the global performance and convergence of the selector using two different use cases. The experimental results demonstrate that the proposed framework allows users enhancing performance while minimizing efforts to tune applications targeted to heterogeneous platforms. Furthermore, we also demonstrate that our framework delivers comparable performance figures with respect to other approaches.