4,443 Hits in 3.1 sec

Massive supercomputing coping with heterogeneity of modern accelerators

Toshio Endo, Satoshi Matsuoka
2008 Proceedings, International Parallel and Distributed Processing Symposium (IPDPS)  
Heterogeneous supercomputers with combined general-purpose and accelerated CPUs promise to be the future major architecture due to their wide-ranging generality and superior performance / power ratio.  ...  For this goal, we divide porting of applications into several steps, analyze performance of the kernel computation, create processes that virtualize the underlying processors, tune parameters with preferences  ...  Despite the fairly long history in supercomputers with vector acceleration options such as the CM-5 and the Meiko CS-2, and the recent high interest on SIMD-vector programming, there have been little results  ... 
doi:10.1109/ipdps.2008.4536251 dblp:conf/ipps/EndoM08 fatcat:vti2acezl5e7tidjcufqkdye7m

The marketplace of high-performance computing

Erich Strohmaier, Jack J Dongarra, Hans W Meuer, Horst D Simon
1999 Parallel Computing  
The initial success of vector computers in the seventies was driven by r a w performance. The introduction of this type of computer systems started the area of 'Supercomputing'.  ...  These criteria determined next to performance the success of MP vector systems especially at industrial customers.  ...  Currently the debate still goes on if we need a new architecture for the very high end supercomputer such as the multithreaded design of Tera.  ... 
doi:10.1016/s0167-8191(99)00067-8 fatcat:wxcevcdcvvdsrcooojjqsmfrfy

The second generation FPS T series: an enhanced parallel vector supercomputer

R. J. Fazzari, J. D. Lynch
1988 Proceedings of the third conference on Hypercube concurrent computers and applications Architecture, software, computer systems, and general issues -  
The FPS T Series is a parallel vector supercomputer incorporating up to 16,384 compute nodes into a network based on the binary n-cube interconnect.  ...  This paper describes the architecture and capabilities of the second generation FPS T Series.  ...  The homogeneous nature of the T Series architecture has facilitated the development of an innovative programming environment for distributed computing [4, 5] .  ... 
doi:10.1145/62297.62306 fatcat:bhwxkjvxfbeoxoi3stqmcbpr2i

The international race towards Exascale in Europe

Fabrizio Gagliardi, Miquel Moreto, Mauro Olivieri, Mateo Valero
2019 CCF Transactions on High Performance Computing  
We cover the political and economic context and make a review of the recent history in high performance computing (HPC) architectures, with special emphasis on the recently announced European initiatives  ...  In this article, we describe the context in which an international race towards Exascale computing has started.  ...  Acknowledgements The authors are grateful to Peter Hsu (Independent consultant) for his valuable technical advice.  ... 
doi:10.1007/s42514-019-00002-y fatcat:kvwcdkzugjclvcyor2pfn7fyve

Main Scientific and Technological Problems in the Field of Architectural Solutions for Supercomputers

Andrey Molyakov
2020 Computer and Information Science  
The author points out the need for a systematic approach, training of young specialists on supporting innovative research.  ...  Moreover, this paper illustrates creation of a domestic processor or processors for solving the problems of creating information systems for processing big data, as well as tasks of artificial intelligence  ...  Do not acknowledge the persons routinely involved in the review and acceptance of manuscripts peer reviewers or editors, associate editors, and consulting editors of the journal in which the article is  ... 
doi:10.5539/cis.v13n3p89 fatcat:qt6k5wpxi5hytbnifubt4l2vca

Hardware and quantum mechanical calculations

E. Wimmer
1992 Philosophical Transactions of the Royal Society of London Series A Physical and Engineering Sciences  
The rem arkable progress in the architecture, speed and capacity of com puter hardw are continues to drive the development of quantum mechanical methods, thus allowing calculations on increasingly complex  ...  The combination of graphics workstations and high-performance supercomputers, integrated in tightly coupled heterogeneous networks, has allowed the design of software systems with unprecedented convenience  ...  The author thanks his former colleagues at Cray Research for many fruitful and stimulating discussions, and especially Charles Grassl for the information on memory component developments.  ... 
doi:10.1098/rsta.1992.0108 fatcat:3e3owgdmbje4linwz4k2nys2q4

Direct numerical simulation of turbulence with a PC/linux cluster

G.-S. Karamanos, C. Evangelinos, R. C. Boes, R. M. Kirby, G. E. Karniadakis
1999 Proceedings of the 1999 ACM/IEEE conference on Supercomputing (CDROM) - Supercomputing '99  
Both low-end and high-end PC clusters, ranging from 2 to 128 processors, are compared to a range of existing supercomputers, such as the IBM SP nodes, Silicon Graphics Origin 2000, Fujitsu AP3000 and Cray  ...  With the rapid development and low cost of PCs, PC clusters are evaluated as a viable low-cost option for scientific computing.  ...  Computations were performed at the following computer centers: Maui High Performance Computing Center (MHPCC), National Center for Supercomputing Applications (NCSA), Major Shared Resource Center of the  ... 
doi:10.1145/331532.331585 dblp:conf/sc/KaramanosEBKK99 fatcat:u27ruwmgzjderaocud5jekh3p4

Optimal selection theory for superconcurrency

R. F. Freund
1989 Proceedings of the 1989 ACM/IEEE conference on Supercomputing - Supercomputing '89  
in part on the bandwidths of die addition the author would like to thank Roxana Kamen storage devices and distributed network used.  ...  be more dilffl'ult to This research was supported by the Office of Na\al break up and assign to different processors and the ability Technology and die Naval Ocean Systems Center-In to do this will rest  ...  ' processors to solve supercomputing architecture.  ... 
doi:10.1145/76263.76342 dblp:conf/sc/Freund89 fatcat:hc67uwdblnbbjhf2r7avrpoisi

A distributed heterogeneous supercomputing management system

A. Ghafoor, J. Yang
1993 Computer  
A.n efficient use of a Distributed Heterogeneous Supercomputing System (DHSS) requhes a thorough understanding of applications and their intelligent scheduling within the system.  ...  An optimd scheduler tries to minimize this cost; the design of which is an NP-complete problem. We describe how network caching can help to reduce the com1)lexity of scheduling in a DHSS.  ...  A DHSS is also expected. to outperform a homogc:neous supercomputing system (HSS) because no matter how powerful a single machine or a set of homogeneous machines might be, HSS cannot satisfy tlhe diverse  ... 
doi:10.1109/2.214443 fatcat:y6r6nqvd65bdzdhvsox2l4c4x4

Future generation supercomputers I

N. Venkateswaran, Arvindakshan Babu, Sudharshan, Deepak Srinivasan, Madhavan Manivannan, T P Ramnath Sai Sagar, Shyamsundar Gopalakrishnan, VinothKrishnan Elangovan, Karthik Chandrasekar, Prem Kumar Ramesh, Viswanath Venkatesan
2007 SIGARCH Computer Architecture News  
We provide a design space based on the proposed model for which a simulator is developed, with the help of which the performance of such a node architecture is outlined.  ...  The potential of such a node architecture can be fully exploited only with an appropriate cluster architecture.  ...  architecture of the node for supercomputing clusters can be fixed as per the computational demands of the application(s).  ... 
doi:10.1145/1360464.1360466 fatcat:zggprey5szg6deoohhxavwjvmu

High-performance Supercomputing as a Risk Evaluation Tool for Geologic Carbon Dioxide Storage

Hajime Yamamoto, Shinichi Nanai, Keni Zhang, Pascal Audigane, Christophe Chiaberge, Ryusei Ogata, Noriaki Nishikawa, Yuichi Hirokawa, Satoru Shingu, Kengo Nakajima
2013 Energy Procedia  
In this study, we implemented TOUGH2-MP code (a parallel version of multi-phase flow simulator TOUGH2) on two different types (vector-and scalar-type) of world-class supercomputers with tens of thousands  ...  In this paper, we present the performances of parallel computation of the code measured on the two supercomputers.  ...  Acknowledgements The use of the Earth Simulator was supported by the "Open Advanced Facilities Initiative for Innovation (Strategic Use by Industry)" funded by the Ministry of Education, Culture, Sports  ... 
doi:10.1016/j.egypro.2013.06.299 fatcat:swvn3qfmwvhdphrcjvqokxuuqm

Accelerating Parallel Maximum Likelihood-Based Phylogenetic Tree Calculations Using Subtree Equality Vectors

A.P. Stamatakis, T. Ludwig, H. Meier, M.J. Wolf
2002 ACM/IEEE SC 2002 Conference (SC'02)  
This paper describes and uses Subtree Equality Vectors (SEVs) to reduce the number of required floating point operations during topology evaluation.  ...  The optimization scales best on clusters of PCs, which also implies a substantial cost saving factor for the determination of large trees. c 0-7695-1524-X/02 $17.00 (c) 2002 IEEE * This work is partially  ...  Furthermore, we would like to thank Ralf Ebner from the LRZ for the technical support and useful information he provided us on the Hitachi SR8000-F1.  ... 
doi:10.1109/sc.2002.10016 dblp:conf/sc/StamatakisLMW02 fatcat:yabjw3yrjfcgtata63bb4h44aq

High Performance Computing with the Cell Broadband Engine

Michael Gschwind, Fred Gustavson, Jan F. Prins
2009 Scientific Programming  
TOP500 list of the world's most powerful supercomputers.  ...  The Cell/B.E. departs from prior architectures by adopting a heterogeneous chip multiprocessor architecture with novel accelerator cores and an explicitly managed memory hierarchy.  ... 
doi:10.1155/2009/979236 fatcat:bqpei34o2jaclhjv3houo5vzb4

Scalable heterogeneous parallelism for atmospheric modeling and simulation

John C. Linford, Adrian Sandu
2010 Journal of Supercomputing  
A function offloading approach is used in a 2D transport module, and a vector stream processing approach is used in a 3D transport module.  ...  This study examines methods for improving the performance of two-dimensional and three-dimensional atmospheric constituent transport simulation on the Cell Broadband Engine Architecture (CBEA).  ...  The authors particularly wish to thank Dr. Willi Homberg of Forschungszentrum Jülich and Prof. Dr. Felix Wolf of the Jülich Supercomputing Centre and RWTH Aachen for their help and support.  ... 
doi:10.1007/s11227-010-0380-8 fatcat:yqynrpo7azewji3in6fdasbmfa

Performance optimizations for scalable CFD applications on hybrid CPU+MIC heterogeneous computing system with millions of cores

Yong-Xian Wang, Li-Lun Zhang, Wei Liu, Xing-Hua Cheng, Yu Zhuang, Anthony T. Chronopoulos
2018 Computers & Fluids  
The tuned solver successfully scales up to half of the entire Tianhe-2 supercomputer system with over 1.376 million of heterogeneous cores.  ...  For computational fluid dynamics (CFD) applications with a large number of grid points/cells, parallel computing is a common efficient strategy to reduce the computational time.  ...  This work was funded by the National Natural Science Foundation of China (NSFC) under grant no. 61379056.  ... 
doi:10.1016/j.compfluid.2018.03.005 fatcat:b7yhhua6tzel5fwphn52ufendy
« Previous Showing results 1 — 15 out of 4,443 results