The GPU performance in coordination of parallel tasks in access to resource groups without conflicts

• Autorzy: Smoliński M.
• Języki publikacji: EN

Abstrakty

EN

In high contention environments, with limited number of shared resources, elimination of resource conflicts between tasks processed in parallel is required. Execution of all tasks without resource conflicts can be achieved by preparing a proper overall schedule for all of them. The effective calculation of conflict-free execution plan for tasks provides the conflictless scheduling algorithm that is dedicated to GPU massively parallel processing. The conflictless scheduling algorithm base on rapid resource conflict detection to mutual exclusion of conflicted tasks in access to global resources and is an alternative to other task synchronization methods. This article presents the performance of modern GPU in calculations of adaptive conflictless task schedule. The performance analysis also takes into account all data transfers to and from the GPU memory in various phases of the conflictless task scheduling algorithm.

Słowa kluczowe

EN

resource conflict elimination; conflict free task execution; mutual exclusion; deadlock avoidance; cooperative concurrency control; GPU massively parallel processing; SIMD control SISD; GPGPU using OpenCL

• Wydawca: Wydawnictwo Szkoły Głównej Gospodarstwa Wiejskiego w Warszawie
• Czasopismo: Information Systems in Management
• Rocznik: 2017
• Tom: Vol. 6, No. 2
• Strony: 143--153
• Opis fizyczny: Bibliogr. 7 poz., rys.

Twórcy

autor

Smoliński M.

• Institute of Information Technology, Lodz University of Technology

Bibliografia

• [1] Amato N., Ravishankar Iyer R., Sundaresan S., and Wu. Y. (1998) A Comparison of Parallel Sorting Algorithms on Different Architectures. Technical Report. Texas A & M University, College Station, TX, USA.
• [2] Flynn M.J., Rudd R. W. (1996) Parallel architectures, ACM Computing Surveys, Volume 28, Issue 1, 67–70
• [3] Martineau M., McIntosh-Smith S., Boulton M., Gaudin W. (2016). An Evaluation of Emerging Many-Core Parallel Programming Models. In Proceedings of the 7th International Workshop on Programming Models and Applications for Multicores and Manycores (PMAM'16), Pavan Balaji and Kai-Cheung Leung (Eds.). ACM, New York, NY, USA
• [4] Smoliński, M. (2016) Coordination of Parallel Tasks in Access to Resource Groups by Adaptive Conflictless Scheduling. Beyond Databases, Architectures and Structures. Advanced Technologies for Data Mining and Knowledge Discovery
• [5] Smoliński, M. (2016) Elimination of task starvation in conflictless scheduling concept. Information Systems in Management Vol. 5, No. 2, 237–247
• [6] Stallings W. (2015) Operating systems, Internals and Design Principles. Pearson Education, 8th edition
• [7] Tanenbaum, A., Bos H. (2014) Modern operating systems. Prentice Hall, 4th edition