ABSTRACT
To sustain performance while facing always tighter power and energy envelopes, High Performance Computing (HPC) is increasingly leveraging heterogeneous architectures. This poses new challenges: to efficiently exploit the available resources, both in terms of hardware and energy, resource management must support a wide range of different heterogeneous devices and programming models that target different application domains. We present a strategy for resource management and programming model support for heterogeneous accelerators for HPC systems with requirements targeting performance, power and predictability. We show how resource management can, in addition to allowing multiple applications to share a set of resources, reduce the burden on the application developer and improve the efficiency of resource allocation.
- ARB 2008. OpenMP Application Program Interface, version 3.0. ARB. http://www.openmp.orgGoogle Scholar
- Cédric Augonnet, Samuel Thibault, Raymond Namyst, and Pierre-André Wacrenier. 2011. StarPU: A Unified Platform for Task Scheduling on Heterogeneous Multicore Architectures. Concurr. Comput.: Pract. Exper. 23, 2 (Feb. 2011), 187--198. Google ScholarDigital Library
- Patrick Bellasi, Giuseppe Massari, and William Fornaciari. 2015. Effective Runtime Resource Management Using Linux Control Groups with the BarbequeRTRM Framework. ACM Trans. Embed. Comput. Syst. 14, 2, Article 39 (March 2015), 17 pages. Google ScholarDigital Library
- José Flich, Giovanni Agosta, Philipp Ampletzer, David Atienza Alonso, Carlo Brandolese, Etienne Cappe, Alessandro Cilardo, Leon Dragić, Alexandre Dray, Alen Duspara, et al. 2017. MANGO: Exploring Manycore Architectures for Next-GeneratiOn HPC Systems. In 2017 Euromicro Conference on Digital System Design (DSD). 478--485.Google Scholar
- Jose Flich, Giovanni Agosta, Philipp Ampletzer, David Atienza Alonso, Alessandro Cilardo, William Fornaciari, Mario Kovac, Fabrice Roudet, and Davide Zoni. 2015. The MANGO FET-HPC Project: An Overview. In Computational Science and Engineering (CSE), 2015 IEEE 18th International Conference on. IEEE, 351--354. Google ScholarDigital Library
- Morris Jette and Mark Grondona. 2003. SLURM: Simple Linux Utility for Resource Management. In ClusterWorld Conference and Expo.Google Scholar
- Khronos OpenCL Working Group. 2014. The OpenCL Specification, Version 1.2. https://www.khronos.org/registry/cl/specs/opencl-1.2.pdf. Aaftab Munshi eds.Google Scholar
- Khronos OpenCL Working Group -- SYCL subgroup. 2014. SYCL™ Specification, Version 1.2. https://www.khronos.org/registry/sycl/specs/sycl-1.2.pdf. Lee Howes and Maria Rovatsou eds.Google Scholar
- Bastian Koller, Nico Struckmann, Jochen Buchholz, and Michael Gienger. 2015. Towards an Environment to Deliver High Performance Computing to Small and Medium Enterprises. Springer International Publishing, Cham, 41--50.Google Scholar
- G. Massari, E. Paone, P. Bellasi, G. Palermo, V. Zaccaria, W. Fornaciari, and C. Silvano. 2014. Combining application adaptivity and system-wide Resource Management on multi-core platforms. In 2014 International Conference on Embedded Computer Systems: Architectures, Modeling, and Simulation (SAMOS XIV). 26--33.Google Scholar
- Microsoft Corporation. 2013. C++ AMP: C++ Accelerated Massive Parallelism, Version 1.2. http://download.microsoft.com/download/4/0/E/40EA02D8-23A7-4BD2-AD3A-0BFFFB640F28/CppAMPLanguageAndProgrammingModel.pdf.Google Scholar
- John Nickolls, Ian Buck, Michael Garland, and Kevin Skadron. 2008. Scalable Parallel Programming with CUDA. ACM Queue 6, 2 (2008), 40--53. Google ScholarDigital Library
- nVidia Corp. 2008. CUDA Technology. http://www.nvidia.com/CUDA. (September 2008).Google Scholar
- A. Pupykina and G. Agosta. 2017. Optimizing Memory Management in Deeply Heterogeneous HPC Accelerators. In 2017 46th International Conference on Parallel Processing Workshops (ICPPW). 291--300.Google Scholar
- Ehsan Totoni, Babak Behzad, Swapnil Ghike, and Josep Torrellas. 2012. Comparing the Power and Performance of Intel's SCC to State-of-the-art CPUs and GPUs. In Proceedings of the 2012 IEEE International Symposium on Performance Analysis of Systems & Software (ISPASS '12). IEEE Computer Society, Washington, DC, USA, 78--87. Google ScholarDigital Library
- Sandra Wienke, Paul Springer, Christian Terboven, and Dieter an Mey. 2012. OpenACC: First Experiences with Real-world Applications. In Proceedings of the 18th International Conference on Parallel Processing (Euro-Par' 12). Springer-Verlag, Berlin, Heidelberg, 859--870. Google ScholarDigital Library
Index Terms
- Managing Heterogeneous Resources in HPC Systems
Recommendations
RADICAL-Pilot and PMIx/PRRTE: Executing Heterogeneous Workloads at Large Scale on Partitioned HPC Resources
Job Scheduling Strategies for Parallel ProcessingAbstractExecution of heterogeneous workflows on high-performance computing (HPC) platforms present unprecedented resource management and execution coordination challenges for runtime systems. Task heterogeneity increases the complexity of resource and ...
Simplifying programming and load balancing of data parallel applications on heterogeneous systems
GPGPU '16: Proceedings of the 9th Annual Workshop on General Purpose Processing using Graphics Processing UnitHeterogeneous architectures have experienced a great development thanks to their excellent cost/performance ratio and low power consumption. But heterogeneity significantly complicates both programming and efficient use of the resources. As a result, ...
Energy Efficient Frequency Scaling on GPUs in Heterogeneous HPC Systems
Architecture of Computing SystemsAbstractWith most major corporations and research institutions having pledged to support sustainability goals for High Performance Computing (HPC), energy efficiency is a critical factor when evaluating heterogeneous HPC systems. However, many popular ...
Comments