OpenCL performance portability for general‐purpose computation on graphics processor units: an exploration on cryptographic primitives

Abstract

Summary The modern trend toward heterogeneous many‐core architectures has led to high architectural diversity in both high performance and high‐end embedded systems. To effectively exploit the computational resources of such a wide range of architectures, programming languages and APIs such as OpenCL have become increasingly popular. Although OpenCL provides functional code portability and the ability to fine tune the application to the target hardware, providing performance portability is still an open problem. Thus, many research works have investigated the optimization of specific combinations of application and target platform. In this paper, we aim at leveraging the experience obtained in the implementation of algorithms from the cryptography domain to provide a set of guidelines for modern many‐core heterogeneous architecture performance portability and to establish a base on which domain‐specific languages and compiler transformations could be built in the near future. We study algorithmic choices and the effect of compiler transformations on three representative applications in the chosen domain on a set of seven target platforms. To estimate how well the application fits the architecture, we define a metric of computational intensity both for the architecture and the application implementation. Besides being useful to compare either different implementation or algorithmic choices and their fitness to a specific architecture, it can also be useful to the compiler to guide the code optimization process. Copyright © 2014 John Wiley & Sons, Ltd.

M2DC – Modular Microserver DataCentre with heterogeneous hardware

“…Regarding security enhancements, in order to provide confidentiality for the transfer of input data and retrieval of the computed results, we will develop efficient OpenCL implementations of cryptographic primitives [11][12][13], which can be transparently deployed either on GPUs or on FPGA targets thanks to the availability of automated tools. The reasons for providing dynamically deployable, yet highly optimized cryptographic primitives are twofold.…”

The ANTAREX tool flow for monitoring and autotuning energy efficient HPC systems

“…The ANTAREX DSL approach aims at reaching a higher abstraction level, to separate and express data communication and computation parallelism, and to augment the capabilities of existing programming models by passing hints and metadata to the compilers for further optimization. The approach aims at improving performance portability with respect to current programming models, such as OpenCL, where fine-tuning of performance (which is very sensitive to even minimal variation in the architectural parameters [10,11]) is left entirely to the programmer. This is done by exploiting the capabilities of the Example of LARA aspect for the inclusion of libVersioningCompiler in a given application.…”

Enabling HPC for QoS-sensitive Applications: The MANGO Approach

“…OpenCL is an open standard for the development of parallel applications on a variety of heterogeneous multi-core architectures [2]. It provides explicit management of heterogeneity, but at a significant cost in terms of tuning performance, which must be performed by the programmer, and building boilerplate code [3], [4]. In MANGO, we aim at integrating the expression of new architectural features as well as QoS concerns and parameters within the existing stack of languages and libraries for extreme-scale HPC systems, by augmenting the runtime library APIs with new functions, as well as by introducing new pragmas or keywords to the language.…”