Geodesic
Development trends of modern supercomputers
Vestnik Ûžno-Uralʹskogo gosudarstvennogo universiteta. Seriâ Vyčislitelʹnaâ matematika i informatika, Tome 8 (2019) no. 3, pp. 92-114 Cet article a éte moissonné depuis la source Math-Net.Ru

Voir la notice de l'article

This work includes analysis of the computation nodes of modern supercomputers from two perspectives; first the hardware components focus and secondly discuss the infrastructure. Identified trends leads to basic options of computation node design. The paper classifies the modern architectures of universal processing cores and specialized hardware accelerators cores; studies the recent trends in memory hierarchy design and intra-node interconnect; the paper includes ways of using the non-volatile memory in modern memory hierarchy. Furthermore, the paper analyses the recent trends in HPC infrastructure, in particular in modern liquid cooling approaches and monitoring. The basic variants of HPC computing nodes design are based on energy efficient universal processor and set of energy-efficient specialized hardware accelerators cores, according to the observed trends. The paper focuses on recent technologies that are currently at various stages of production or at the functional prototype stage. The study also discusses state-of-the-art computational challenges and algorithms-to-architecture mapping issues. Lastly, the paper discusses the current technological problems and main areas to maintain the progress in HPC area.
Keywords: high-performance compute node, architecture development of a high-performance compute node, analysis of supercomputer architectures, development trends of supercomputers. 
@article{VYURV_2019_8_3_a4,
     author = {E. {\CYRO}. Tyutlyaeva and I. O. Odintsov and A. A. Moskovsky and G. V. Marmuzov},
     title = {Development trends of modern supercomputers},
     journal = {Vestnik \^U\v{z}no-Uralʹskogo gosudarstvennogo universiteta. Seri\^a Vy\v{c}islitelʹna\^a matematika i informatika},
     pages = {92--114},
     year = {2019},
     volume = {8},
     number = {3},
     language = {ru},
     url = {http://geodesic.mathdoc.fr/item/VYURV_2019_8_3_a4/}
}
TY  - JOUR
AU  - E. О. Tyutlyaeva
AU  - I. O. Odintsov
AU  - A. A. Moskovsky
AU  - G. V. Marmuzov
TI  - Development trends of modern supercomputers
JO  - Vestnik Ûžno-Uralʹskogo gosudarstvennogo universiteta. Seriâ Vyčislitelʹnaâ matematika i informatika
PY  - 2019
SP  - 92
EP  - 114
VL  - 8
IS  - 3
UR  - http://geodesic.mathdoc.fr/item/VYURV_2019_8_3_a4/
LA  - ru
ID  - VYURV_2019_8_3_a4
ER  - 
%0 Journal Article
%A E. О. Tyutlyaeva
%A I. O. Odintsov
%A A. A. Moskovsky
%A G. V. Marmuzov
%T Development trends of modern supercomputers
%J Vestnik Ûžno-Uralʹskogo gosudarstvennogo universiteta. Seriâ Vyčislitelʹnaâ matematika i informatika
%D 2019
%P 92-114
%V 8
%N 3
%U http://geodesic.mathdoc.fr/item/VYURV_2019_8_3_a4/
%G ru
%F VYURV_2019_8_3_a4
E. О. Tyutlyaeva; I. O. Odintsov; A. A. Moskovsky; G. V. Marmuzov. Development trends of modern supercomputers. Vestnik Ûžno-Uralʹskogo gosudarstvennogo universiteta. Seriâ Vyčislitelʹnaâ matematika i informatika, Tome 8 (2019) no. 3, pp. 92-114. http://geodesic.mathdoc.fr/item/VYURV_2019_8_3_a4/

[1] The project of the Russian Academy of Science: «Development of the compute system for simulation of the exascale supercomputer» } {\tt http://www.keldysh.ru/projects/exaflops.pdf

[2] D. A. Reed, J. Dongarra, “Exascale Computing and Big Data”, Communications of the ACM, 58:7 (2015), 56–68 | DOI

[3] G. Chrysos, “Intel Xeon Phi coprocessor (codename Knights Corner)”, Proceedings of the 2012 IEEE Hot Chips 24 Symposium (HCS, August 27–29, 2012, Cupertino, CA), 2012, 1–31 | DOI

[4] E. N. Lindholm, S. Oberman, J. Montrym, “NVIDIA Tesla: A Unified Graphics and Computing Architecture”, IEEE Micro, 28:2 (2008), 39–55 | DOI

[5] N. Jouppi, C. Young, N. Patil, D. Patterson, “Motivation for and Evaluation of the First Tensor Processing Unit”, IEEE Micro, 38:3 (2018), 10–19 | DOI

[6] M. Davies, et al., “Loihi: A Neuromorphic Manycore Processor with On-Chip Learning”, IEEE Micro, 38:1 (2018), 82–99 | DOI

[7] J. Hsu, CES 2018: Intel’s 49-Qubit Chip Shoots for Quantum Supremacy. IEEE Spectrum Tech Talks, 2018 } {\tt https://spectrum.ieee.org/tech-talk/

[8] Intel Stratix 10 SoC FPGAs } {\tt https://www.intel.com/content/www/us/en/

[9] Exascale Requirements Review. An Office of Science review sponsored jointly by Advanced Scientific Computing Research and High Energy Physics. } {\tt http://hepcce.org/files/2016/11/

[10] Top500 List Statistics. Release November 2018 } {\tt https://www.top500.org/statistics/list/

[11] N. Hemsoth, Cascade Lake at Heart of 2019 TACC Supercomputer. Onlain resurs tekhnologicheskikh novostei Next Platform podderzhivaemyi Stackhouse Publishing Inc v partnerstve s The Register, 2018 } {\tt https://www.nextplatform.com/2018/08/29/

[12] V. Bartsch, D6.3 Initial Project Press Release. ExaNoDe Consortium Public deliverable. Press-reliz po proektu ExaNode, 2016 } {\tt http://exanode.eu/wp-content/

[13] ARMv8 — A Scalable Vector Extension for Post-K. FUJITSU LIMITED, 2016 } {\tt http://www.fujitsu.com/global/Images/

[14] Astra. Top500 The List } {\tt https://www.top500.org/system/179565

[15] Xilinx. High Performance Computing and Data Storage } {\tt https://www.xilinx.com/applications/high-performance-computing.html

[16] A. N. Timmel, J. T. Daly, Multiplication with Fourier Optics Simulating 16-bit Modular Multiplication } {\tt https://arxiv.org/pdf/1801.01121.pdf

[17] A. K. Kim, V. I. Perekatov, V. M. Feldman, “On the way to russian exasistemes: plans of the Elbrus hardware-software platform develоpers on creation of an exaflops performance supercomputer”, Voprosy radioelektroniki, 2018, no. 2, 6–13

[18] CORAL Collaboration: Briefing on CORAL-2 RFP and Draft Technical Requirements } {\tt https://procurement.ornl.gov/

[19] R. Farber, HPC and AI — Two Communities Same Future. HPCwire: Global News and Information on High Performance Computing, 2018 } {\tt https://www.hpcwire.com/2018/01/25/

[20] JEDEC DDR5 NVDIMM-P Standards Under Development. Global Standards for the Microelectronics Industry, 2017 } {\tt https://www.jedec.org/news/

[21] R. Hadidi, et al., “Demystifying the Characteristics of 3D-Stacked Memories: A Case Study for Hybrid Memory Cube”, Proceedings of the IEEE International Symposium on Workload Characterization (IISWC 2017, October 1–3, 2017, Seattle, WA, USA), 66–75 | DOI

[22] High Bandwidth Memory (HBM) DRAM JESD235A. Global Standards for the Microelectronics Industry, 2015 } {\tt https://www.jedec.org/

[23] Hybrid Memory Cube (HMC). Hybrid Memory Cube Consortium Page } {\tt http://hybridmemorycube.org/

[24] Intel Memory Drive Technology Application Note } {\tt https://www.intel.com/content/dam/support/us/en/

[25] Graphics Double Data Rate (GDDR5) SGRAM standard. JESD212C. Global Standards for the Microelectronics Industry., 2016 } {\tt https://www.jedec.org/

[26] Graphics Double Data Rate (GDDR6) SGRAM standard. JESD250A. Global Standards for the Microelectronics Industry., 2017 } {\tt https://www.jedec.org/

[27] da Silva R. Ferreira, S. Callaghan, E. Deelman, “On the use of burst buffers for accelerating data-intensive scientific workflows”, Proceedings of the 12th Workshop on Workflows in Support of Large-Scale Science (WORKS ’17), ACM, 2017, 1–9 | DOI

[28] W. Bhimji, D. Bard, M. Romanus, D. Paul, A. Ovsyannikov, B. Friesen, et al., Accelerating Science with the NERSC Burst Buffer Early User Program. Lawrence Berkeley National Laboratory., 2016 } {\tt https://escholarship.org/uc/

[29] CrayDataWarpTM Applications I/O Accelerator } {\tt https://www.cray.com/products/storage/datawarp

[30] T. P. Morgan, For many hyperconverged is the next platform. Onlain resurs tekhnologicheskikh novostei Next Platform podderzhivaemyi Stackhouse Publishing Inc v partnerstve s The Register, 2018 } {\tt https://www.nextplatform.com/2018/01/29/

[31] V. Bahur, The RSC Technologies Company introduces hyperconverged HPC-solution based on state-of-the-are components., 2018 } {\tt http://www.cnews.ru/news/

[32] The GEN-Z Consortium } {\tt https://genzconsortium.org/

[33] The OpenCAPI Consortium } {\tt https://opencapi.org/

[34] NVLink Fabric } {\tt https://www.nvidia.com/ru-ru/data-center/nvlink/

[35] Product Brief: Intel Xeon Scalable Platform. } {\tt https://www.intel.ru/content/www/ru/ru/processors/xeon/scalable/

[36] Infinity Fabric (IF) — AMD } {\tt https://en.wikichip.org/wiki/amd/

[37] Russian microprocessors «Elbrus» and «MCST R» series and boards based thereon. Production catalogue 2017. } {\tt http://mcst.ru/files/

[38] CCIX } {\tt https://www.ccixconsortium.com/

[39] Coherent Accelerator Processor Interface (CAPI) } {\tt https://developer.ibm.com/linuxonpower/capi/

[40] V. Shustikov, Skoltex researches developed supercomputer «Zhores». The Skolkovo Foundation Press-Release, 2019 \href{https://sk.ru/news/b/pressreleases/archive/2019/01/18/}{{\tt https://sk.ru/news/b/pressre

[41] Is Liquid Cooling Ready to Go Mainstream? HPCwire: Global News and Information on High Performance Computing } {\tt https://www.hpcwire.com/2017/02/13/

[42] Aquila } {\tt https://www.aquilagroup.com/cooling/

[43] RSC Group. } {\tt http://www.rscgroup.ru/en/

[44] Asetek } {\tt https://www.asetek.com/

[45] Ebullient } {\tt http://ebullientcooling.com/

[46] ExaScaler Inc. Overview } {\tt http://www.exascaler.co.jp/en/company

[47] 3M Server Solutions for Data Centers } {\tt https://www.3m.com/3M/

[48] S.\M Abramov, S. A. Amelkin, A. Yu. Romanenko, A. S. Simonov, A. A. Chichkovskii, “Opyt realizatsii vysokoproizvoditelnykh vychislitelnykh sistem s pogruzhnoi zhidkostnoi sistemoi okhlazhdeniya”, Trudy 3-i Vserossiiskoi nauchno-tekhnicheskoi konferentsii «Superkompyuternye tekhnologii» (SKT-2014) (Divnomorskoe, Gelendzhik, 29 sentyabrya–4 oktyabrya 2014 g), 2014, 9–15

[49] I. I. Levin, A. I. Dordopulo, Yu. I. Doronchenko, M. K. Raskladkin, A. M. Fedorov, Pogruzhnaya sistema okhlazhdeniya rekonfiguriruemykh vychislitelnykh sistem na osnove PLIS. Programmnye sistemy: teoriya i prilozheniya. 2016. #4 (31) } {\tt https://cyberleninka.ru/

[50] Liquid MIPS } {\tt http://www.liquidmips.com/cms/en-us/

[51] A. Libri, A. Bartolini, L. Benini, Dwarf in a Giant: Enabling Scalable, High-Resolution HPC Energy Monitoring for Real-Time Profiling and Analytics } {\tt https://arxiv.org/pdf/1806.02698.pdf

[52] R. E. Grant, M. Levenhagen, S. L. Olivier, D. DeBonis, K. T. Pedretti, I. I. Laros, “Standardizing Power Monitoring and Control at Exascale”, Computer, 49:10 (2016), 38–46 | DOI

[53] Y. Georgiou, D. Glesser, D. Trystram, “Adaptive Resource and Job Management for Limited Power Consumption”, IEEE International Parallel and Distributed Processing Symposium Workshop (2015, Hyderabad), 863–870 | DOI

[54] J. M. Shalf, R. Leland, “Computing beyond Moore’s Law”, Computer, 48:12 (2015), 14–23 | DOI