ABSTRACT
The paper focuses on the latest trends in supercomputer and high-performance computer architectures with emphasis of the important parameters for parallel computer performance. The main objective is to propose the high-speed network architecture for collective communications in supercomputers "Silicon Galaxy" that provides high-bandwidth, low latency and high tolerance to failures. The proposed network architecture "Silicon Galaxy" is composed of "Milky Way Galaxy" network for intercluster communication and "Dwarf Galaxy" network for communication within the cluster segments (intracluster communication in supercomputers). Communication performance of the proposed collective network has been evaluated on the base of parallel computer simulations on IBM Blade Center.
- Abad Pablo, Prieto Pablo, Menezo Lucía, Colaso Adrián, Puente Valentin, Gregorio José-Ángel, TOPAZ: An Open-Source Interconnection Network Simulator for Chip Multiprocessors and Supercomputers, 2012 Sixth IEEE/ACM International Symposium on Networks-on-Chip, 978-0-7695-4677-3, pp. 99--106. Google ScholarDigital Library
- Borovska Pl, Ivanova D., Architectural Design of Grand Clos Collective Network for Supercomputers, Proceedings of the 17th International Conference on Computers, Recent Advances in Computer Science, WSEAS, Rhodes Island, Greece July 16--19, 2013, ISBN: 978-960-474-311-7, pp. 146--151.Google Scholar
- Borovska Pl., Nakov O., Ivanova D., Ivanov K., Georgiev G., Communication Performance, Evaluation and Analysis of a Mesh System Area Network for High Performance Computers, 12-th WSEAS International Conference on Mathematical Methods, Computational Techniques and Intelligence Systems (MAMECTIS'10), Kantaoui, Sousse, Tunisia, May 3--6, 2010, ISBN: 978-960-474-188-5, pp. 217--222. Google ScholarDigital Library
- Fraigniaud Pierre, Phan Hoang-Anh, Degree hunter: on the impact of balancing node degrees in de Bruijn-based overlay networks, IEEE International Symposium on Parallel and Distributed Processing Workshops and PhD Forum, 2011 I 978-1-4244-6534-7/10, pp. 1--8.Google Scholar
- Hoefler Torsten, Schneider Timo, Optimization principles for collective neighborhood communications, International Conference for High Performance Computing, Networking, Storage and Analysis, 2012, ISBN: 978-1-4673-0805-2, pp. 1--10. Google ScholarDigital Library
- Ivanova D., System architecture network for collective communication "GRAND-CLOS" in supercomputers, National Youth School for Scientific and Exchange Ideas in Information and Communication Technologies, Rousse 2013, pp. 67--72.Google Scholar
- Pilla Laercio, Ribeiro Christiane Pousa, Cordeiro Daniel, Mei Chao, Abhinav Bhatele, Philippe O. A. Navaux, Francois Broquedis, Jean-Francois Mehaut, Laxmikant V. Kale, A Hierarchical Approach for Load Balancing on Parallel Multi-core Systems, 2012 41st International Conference on Parallel Processing (ICPP), 0190-3918/12, pp. 118--127. Google ScholarDigital Library
Index Terms
- "Silicon Galaxy" system area network for collective communication in supercomputers
Recommendations
Evaluating MPI Collective Communication on the SP2, T3D, and Pargon Multicomputers
HPCA '97: Proceedings of the 3rd IEEE Symposium on High-Performance Computer ArchitectureWe evaluate the architectural support of collective communication operations on the IBM SP2,Cray T3D, and Intel Paragon. The MPI performance data are obtained from the STAP benchmark experiments jointly performed at the USC and HKU. The T3D demonstrated ...
Performance characteristics of hybrid MPI/OpenMP implementations of NAS parallel benchmarks SP and BT on large-scale multicore supercomputers
Special issue on the 1st international workshop on performance modeling, benchmarking and simulation of high performance computing systems (PMBS 10)The NAS Parallel Benchmarks (NPB) are well-known applications with the fixed algorithms for evaluating parallel systems and tools. Multicore supercomputers provide a natural programming paradigm for hybrid programs, whereby OpenMP can be used with the ...
Communication performance of a recirculative omega high-speed system area network for HPC
CompSysTech '11: Proceedings of the 12th International Conference on Computer Systems and TechnologiesAs modern scientific and engineering computing matures, the demands for high-speed computational resources are growing at a rapid rate. The communication performance of system area network is a crucial factor influencing the parallel performance of ...
Comments