SafeGPU: Contract- and library-based GPGPU for object-oriented languages. (June 2017)

• Record Type: Journal Article
• Authors: Kolesnichenko, Alexey; Poskitt, Christopher M.; Nanz, Sebastian
• Is Part Of: Computer languages, systems & structures. Volume 48(2017)
• Page Start: 68
• Page End: 88
• Publication Date: 2017-06
• Subjects: GPGPU -- Parallel computing -- Runtime code generation -- Generative programming -- Object-orientation -- Managed languages -- Design-by-contract -- Program correctness; Programming languages (Electronic computers) -- Periodicals; Computer networks -- Periodicals; Computer architecture -- Periodicals; Computer systems -- Periodicals; Langage de programmation; Réseau d'ordinateurs; Architecture d'ordinateur; Périodique électronique (Descripteur de forme); Ressource Internet (Descripteur de forme); 005.13
• DOI: 10.1016/j.cl.2016.08.002
• Languages: English
• ISSNs: 1477-8424

Abstract

Abstract: Using GPUs as general-purpose processors has revolutionized parallel computing by providing, for a large and growing set of algorithms, massive data-parallelization on desktop machines. An obstacle to their widespread adoption, however, is the difficulty of programming them and the low-level control of the hardware required to achieve good performance. This paper proposes a programming approach, SafeGPU, that aims to make GPU data-parallel operations accessible through high-level libraries for object-oriented languages, while maintaining the performance benefits of lower-level code. The approach provides data-parallel operations for collections that can be chained and combined to express compound computations, with data synchronization and device management all handled automatically. It also integrates the design-by-contract methodology, which increases confidence in functional program correctness by embedding executable specifications into the program text. We present a prototype of SafeGPU for Eiffel, and show that it leads to modular and concise code that is accessible for GPGPU non-experts, while still providing performance comparable with that of hand-written CUDA code. We also describe our first steps towards porting it to C#, highlighting some challenges, solutions, and insights for implementing the approach in different managed languages. Finally, we show that runtime contract-checking becomes feasible in SafeGPU, as the contracts can be executed on the GPU. Abstract : Highlights: A library-based GPU programming approach for object-oriented languages; implemented for Eiffel, and partially ported to C#. Supports modular, efficient, and customizable GPU programming. Automatically generates and optimizes CUDA kernels from high-level code. Handles the transfer of primitives, simple classes, and function abstractions to C++ and CUDA. Supports contracts and efficient runtime assertion checking on the GPU.