Extending Amdahl's and Gustafson Baris's Laws by Adding Communication Overheads
Main Article Content
Abstract
The extension of Amdahl's law and Gustafson-Barsis' law presented in this article provides insight into the relationship between communication costs and network topology in parallel computing. By considering communication latency in parallel computation execution time, these extensions can help researchers and developers optimize the interconnected network architecture of processing nodes and improve the performance of parallel computing systems. Today, parallel computer systems consisting of hundreds and thousands of processing nodes based on multiprocessor chip technology, high-speed optical transmission such as supercomputers are being researched, developed, and applied in many fields. many areas. Although chip technology has progressed to the 3nm process, the network architecture connecting processing nodes continues to be a problem that greatly affects the communication delay in the parallel computation time of the application problems. use. In this paper, extensions of Amdahl's law and Gustafson-Barsis' law are presented with the addition of communication costs depending on the topology of the topology. These extensions provide insight into the relationship between communication costs and network topology in parallel computing. By considering communication latency in parallel computation execution time, these extensions can help researchers and developers optimize the interconnected network architecture of processing nodes and improve the performance of parallel computing systems.
Keywords
Extensions of Amdahl’Law, Gustafson-Barsis, interconnection network topology, supercomputer, communication overhead, speedup.
Article Details
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
References
[1] GREEN500 LIST (accessed in June, 2022). Online Avaiable: https://www.top500.org/lists/green500/list/2022/06/.
[2] Anthony Spadafora, Next-gen Nvidia Mellanox InfiniBand will take supercomputers to the next level. Online Avaiable: https://www.techradar.com/news/next-gen-nvidiamellanox-infiniband-will-take-supercomputers-to-thenext-level. published November 19, 2020.
[3] InfiniBand Accelerates Six of the Top Ten Supercomputers in the World, Including the Top Three, and Four of the Top Five on June's TOP500 (accessed in June 17, 2019). Online Avaiable: https://nvidianews.nvidia.com/news/infinibandaccelerates-six-of-the-top-ten-supercomputers-in-theworld-including-the-top-three-and-four-of-the-topfive-on-june-s-top500
[4] T. Zidenberg and I. Keslassy and U. Weiser, MultiAmdahl: How Should I Divide My Heterogenous Chip? IEEE Computer Architecture Letters, Volume: 11, Issue: 2, July-Dec. 2012. https://doi.org/10.1109/L-CA.2012.3
[5] János Végh, How Amdahl’s low restricts supercomputer applications and building ever bigger supercomputers. University of Miskolc, Hungary. [v1] Fri, 4 Aug 2017 11:56:45 UTC (44 KB). [v2] Fri, 29 Dec 2017 06:26:45 UTC (33 KB) [6] Uri Verner, Avi Mendelson, and Assaf Schuster, Extending Amdahl’s Law for Multicores with Turbo Boost. Dept. of Computer Science, Technion, Israel. Published 2017. Computer Science IEEE Computer Architecture Letters. https://doi.org/10.1109/LCA.2015.2512982
[7] S. Pei, M-Seo Kim, and J. Luc Gaudiot, Extending Amdahl's Law for Heterogeneous Multicore Processor with Consideration of the Overhead of Data Preparation. January 2016. IEEE Embedded Systems Letters 8(1) https://doi.org/10.1109/LES.2016.2519521
[8] Chaitanya Poolla, Rahul Saxena, On extending Amdahl’s law to learn computer performance. Microprocessors and Microsystems. Volume 96, February 2023, 104745. © 2022. Elsevier B.V. All rights reserved. https://doi.org/10.1016/j.micpro.2022.104745
[9] Ami Marowka, Extending Amdahl's Law for Heterogeneous Computing. Conference: Parallel and Distributed Processing with Applications (ISPA), 2012 IEEE 10th International Symposium on. https://doi.org/10.1109/ISPA.2012.47 [10] Yuichiro Ajima, The Tofu Interconnect D for Supercomputer,. Fujitsu Limited, June 20 2019
[11] Gusfson John L, Gustafson’s Law. All content following this page was uploaded by John Gustafson on 24 January 2022.
[2] Anthony Spadafora, Next-gen Nvidia Mellanox InfiniBand will take supercomputers to the next level. Online Avaiable: https://www.techradar.com/news/next-gen-nvidiamellanox-infiniband-will-take-supercomputers-to-thenext-level. published November 19, 2020.
[3] InfiniBand Accelerates Six of the Top Ten Supercomputers in the World, Including the Top Three, and Four of the Top Five on June's TOP500 (accessed in June 17, 2019). Online Avaiable: https://nvidianews.nvidia.com/news/infinibandaccelerates-six-of-the-top-ten-supercomputers-in-theworld-including-the-top-three-and-four-of-the-topfive-on-june-s-top500
[4] T. Zidenberg and I. Keslassy and U. Weiser, MultiAmdahl: How Should I Divide My Heterogenous Chip? IEEE Computer Architecture Letters, Volume: 11, Issue: 2, July-Dec. 2012. https://doi.org/10.1109/L-CA.2012.3
[5] János Végh, How Amdahl’s low restricts supercomputer applications and building ever bigger supercomputers. University of Miskolc, Hungary. [v1] Fri, 4 Aug 2017 11:56:45 UTC (44 KB). [v2] Fri, 29 Dec 2017 06:26:45 UTC (33 KB) [6] Uri Verner, Avi Mendelson, and Assaf Schuster, Extending Amdahl’s Law for Multicores with Turbo Boost. Dept. of Computer Science, Technion, Israel. Published 2017. Computer Science IEEE Computer Architecture Letters. https://doi.org/10.1109/LCA.2015.2512982
[7] S. Pei, M-Seo Kim, and J. Luc Gaudiot, Extending Amdahl's Law for Heterogeneous Multicore Processor with Consideration of the Overhead of Data Preparation. January 2016. IEEE Embedded Systems Letters 8(1) https://doi.org/10.1109/LES.2016.2519521
[8] Chaitanya Poolla, Rahul Saxena, On extending Amdahl’s law to learn computer performance. Microprocessors and Microsystems. Volume 96, February 2023, 104745. © 2022. Elsevier B.V. All rights reserved. https://doi.org/10.1016/j.micpro.2022.104745
[9] Ami Marowka, Extending Amdahl's Law for Heterogeneous Computing. Conference: Parallel and Distributed Processing with Applications (ISPA), 2012 IEEE 10th International Symposium on. https://doi.org/10.1109/ISPA.2012.47 [10] Yuichiro Ajima, The Tofu Interconnect D for Supercomputer,. Fujitsu Limited, June 20 2019
[11] Gusfson John L, Gustafson’s Law. All content following this page was uploaded by John Gustafson on 24 January 2022.