Jump to content

TOP500

From Wikipedia, the free encyclopedia

This is an old revision of this page, as edited by 89.204.135.158 (talk) at 20:56, 4 December 2016 (→Systems ranked #1: +Cray). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.

Distribution of supercomputers in the TOP500 list by country

The TOP500 project ranks and details the 500 most powerful non-distributed computer systems in the world. The project was started in 1993 and publishes an updated list of the supercomputers twice a year. The first of these updates always coincides with the International Supercomputing Conference in June, and the second is presented in November at the ACM/IEEE Supercomputing Conference. The project aims to provide a reliable basis for tracking and detecting trends in high-performance computing and bases rankings on HPL,[1] a portable implementation of the high-performance LINPACK benchmark written in Fortran for distributed-memory computers. Currently, the China made Sunway TaihuLight in the People's Republic of China (PRC) is the world's most powerful and fastest supercomputer, reaching 93.015 petaflops on the LINPACK benchmarks.

The TOP500 list is compiled by Jack Dongarra of the University of Tennessee, Knoxville, Erich Strohmaier and Horst Simon of NERSC/Lawrence Berkeley National Laboratory (and, from 1993 until his death in 2014, Hans Meuer of the University of Mannheim, Germany.)

History

Rapid growth of supercomputer performance, based on data from top500.org site. The logarithmic y-axis shows performance in GFLOPS.
  Combined performance of 500 largest supercomputers
  Fastest supercomputer
  Supercomputer in 500th place

In the early 1990s, a new definition of supercomputer was needed to produce meaningful statistics. After experimenting with metrics based on processor count in 1992, the idea was born at the University of Mannheim to use a detailed listing of installed systems as the basis. In early 1993, Jack Dongarra was persuaded to join the project with his LINPACK benchmark. A first test version was produced in May 1993, partially based on data available on the Internet, including the following sources:[2][3]

The information from those sources was used for the first two lists. Since June 1993, the TOP500 is produced bi-annually based on site and vendor submissions only.

Since 1993, performance of the #1 ranked position has steadily grown in agreement with Moore's law, doubling roughly every 14 months. Tianhe-2 (that was fastest in November 2014) with a Rpeak[6] of 54.9024 PFLOPS, is over 419,102 times faster than the fastest system in November 1993, the Connection Machine CM-5/1024 (1024 cores) with Rpeak of 131.0 GFLOPS.[7]

Architecture and operating systems

In June 2016, a Chinese computer made the top based on SW26010 processors, a new, radically modified, model in the Sunway (or ShenWei) line.

As of November 2016, TOP500 supercomputers are now all 64-bit, mostly based on x86-64 CPUs (Intel EMT64 and AMD AMD64 instruction set architecture), with few exceptions (all RISC-based) including 22 supercomputers based on Power Architecture used by IBM POWER microprocessors, seven SPARC (all Fujitsu/SPARC-based, one of which surprisingly made the top in 2011 without a GPU, currently ranked seventh), and two, seemingly related, Chinese designs: the ShenWei-based (ranked 11 in 2011, ranked 158th in November 2016) and Sunway SW26010-based ranked 1 in 2016, making up the remainder (another non-US design is PEZY-SC, while it's an accelerator paired with Intel's Xeon). Prior to the ascendance of 32-bit x86 and later 64-bit x86-64 in the early 2000s, a variety of RISC processor families made up the majority of TOP500 supercomputers, including RISC architectures such as SPARC, MIPS, PA-RISC and Alpha.

In recent years heterogeneous computing, mostly using Nvidia's graphics processing units (GPU) as coprocessors, has become a popular way to reach a better performance per watt ratio and higher absolute performance; it is almost required for good performance and to make the top (or top 10), with some exceptions, such as the mentioned SPARC computer without any coprocessors. An x86-based coprocessor, Xeon Phi, has also been used.

Share of processor architecture families in TOP500 supercomputers by time trend

All the fastest supercomputers in the decade since the Earth Simulator supercomputer have used a Linux-based operating system. As of November 2016, 498 or 99.6% of the world's fastest supercomputers use the Linux kernel. The remaining two or 0.4%, run AIX operating system, a variant of Unix.[8] Within those 99.6% running Linux, are the most powerful supercomputers including all those ranking as the top ten (and virtually all the fastest computers).

The non-Linux computers on the list – the two AIX ones – run on POWER7 (ranked 386th and 387th[9]). Those are made by IBM. IBM has higher ranked computers running Linux, including the 4th and 6th.

Since November 2015, no computer on the list runs Windows. In November 2014, Windows Azure[10] cloud computer was no longer on the list of fastest supercomputers (its best rank was 165 in 2012), leaving the Shanghai Supercomputer Center's "Magic Cube" as the only Windows-based supercomputer on the list, until "Magic Cube" also dropped off the list. "Magic Cube" was ranked 436 in its last appearance on the list released in June 2015 (its best rank was 11th in 2008).[11]

Top 10 ranking

Top 10 positions of the 48th TOP500 in November 2016[12]
Rank Rmax
Rpeak
(PFLOPS) 		Name Model Processor Interconnect Vendor Site
Country, year 		Operating system
1 93.015
125.436 		Sunway TaihuLight Sunway MPP SW26010 Sunway[13] NRCPC National Supercomputing Center in Wuxi
  China, 2016[13] 		Linux (Raise)
2 33.863
54.902 		Tianhe-2 TH-IVB-FEP Xeon E5–2692, Xeon Phi 31S1P TH Express-2 NUDT National Supercomputing Center in Guangzhou
  China, 2013 		Linux (Kylin)
3 17.590
27.113 		Titan Cray XK7 Opteron 6274, Tesla K20X Gemini Cray Inc. Oak Ridge National Laboratory
  United States, 2012 		Linux (CLE, SLES based)
4 17.173
20.133 		Sequoia Blue Gene/Q PowerPC A2 Custom IBM Lawrence Livermore National Laboratory
  United States, 2013 		Linux (RHEL and CNK)
5 14.015
27.881 		Cori Cray XC40 Xeon Phi 7250 Aries Cray Inc. National Energy Research Scientific Computing Center
  United States, 2016 		Linux (CLE)
6 13.555
24.914 		Oakforest-PACS Fujitsu Xeon Phi 7250 Intel Omni-Path Fujitsu Joint Center for Advanced High Performance Computing, Kashiwa
  Japan, 2016 		Linux
7 10.510
11.280 		K computer Fujitsu SPARC64 VIIIfx Tofu Fujitsu RIKEN Advanced Institute for Computational Science (AICS)
 Japan, 2011 		Linux
8 9.779
15.988 		Piz Daint Cray XC50 Xeon E5-2690v3, Tesla P100 Aries Cray Inc. Swiss National Supercomputing Centre
  Switzerland, 2016 		Linux (CLE)
9 8.587
10.066 		Mira Blue Gene/Q PowerPC A2 Custom IBM Argonne National Laboratory
 United States, 2012 		Linux (CNK)
10 8.101
11.079 		Trinity Cray XC40 Xeon E5–2698v3 Aries Cray Inc. Los Alamos National Laboratory
  United States, 2015 		Linux (CLE)

Legend:

  • Rank – Position within the TOP500 ranking. In the TOP500 list table, the computers are ordered first by their Rmax value. In the case of equal performances (Rmax value) for different computers, the order is by Rpeak. For sites that have the same computer, the order is by memory size and then alphabetically.
  • Rmax – The highest score measured using the LINPACK benchmark suite. This is the number that is used to rank the computers. Measured in quadrillions of floating point operations per second, i.e. petaflops.
  • Rpeak – This is the theoretical peak performance of the system. Measured in PFLOPS.
  • Name – Some supercomputers are unique, at least on its location, and are therefore named by their owner.
  • Model – The computing platform as it is marketed.
  • Processor cores – The number of active processor cores actively used running LINPACK. After this figure is the processor architecture of the cores named. If the interconnect between computing nodes is of interest, it's also included here.
  • Vendor – The manufacturer of the platform and hardware.
  • Site – The name of the facility operating the supercomputer.
  • Country – The country in which the computer is situated.
  • Year – The year of installation/last major update.
  • Operating system – The operating system that the computer uses.

Other rankings

Top countries

Numbers below represent the number of computers in the TOP500 that are in each of the listed countries.

Systems ranked #1 since 1976

Number of systems

By number of systems as of June 2016:[17]

Template:Multicol

Top five processor generations
Top five vendors

Template:Multicol-break

Top five operating systems

Note all of the above are Linux, just the top one is "generic" Linux.

Operating system family

Template:Multicol-end

New developments in supercomputing

In November 2014, it was announced that the United States was developing two new supercomputers to dethrone China's Tianhe-2 from its position as world's fastest supercomputer. The two computers, Sierra and Summit, will each exceed Tianhe-2's 55 peak petaflops. Summit, the more powerful of the two, will deliver 150–300 peak petaflops.[19] On 10 April 2015, US government agencies banned Nvidia[20] and Intel Corporation from providing Xeon chips to China due to their use, according to the US, in researching nuclear weapons – research which US law bans US companies from contributing towards.[21]

On 29 July 2015, President Obama signed an executive order creating a National Strategic Computing Initiative calling for the accelerated development of an exascale (1000 petaflop) system and funding research into post-semiconductor computing.[22]

In June 2016, Japaneses Fujitsu announced at the International Supercomputing Conference that its future exascale supercomputer will feature processors of its own design that implement the ARMv8 architecture. The Flagship2020 program, by Fujitsu for RIKEN plans to break the exaflops barrier by 2020 (and "it looks like China and France have a chance to do so and that the United States is content – for the moment at least – to wait until 2023 to break through the exaflops barrier."[23]) These processors will also implement extensions to the ARMv8 architecture equivalent to HPC-ACE2 that Fujitsu is developing with ARM Holdings.[23]

Large machines not on the list

Some major systems are not listed on the list. The largest example is the NCSA's Blue Waters which publicly announced the decision not to participate in the list[24] because they do not feel it accurately indicate the ability for any system to be able to do useful work.[25] Other organizations decide not to list systems for security and/or commercial competitiveness reasons. Additional purpose-built machines that are not capable or do not run the benchmark were not included, such as RIKEN MDGRAPE-3 and MDGRAPE-4.

Computers and architectures that drop off the list

IBM Roadrunner[26] is no longer on the list (or any other using the Cell coprocessor, or PowerXCell as in the Roadrunner supercomputer), but it is an example of a computer that would easily be included, if it had not been decommissioned, as it is faster than the one ranked 500th.[27]

Conversely, computers, such as the Microsoft Azure,[28] have dropped off the list simply because the stated performance numbers are no longer high enough, while theoretically the computers could have been upgraded to get faster (or not) without being reported.

All Itanium based systems (including the one which reached second rank in 2004[29])[30] and (non-SIMD-style) vector processors (NEC-based such as the Earth simulator that was fastest in 2002[31]) have also fallen off the list. Similarly the Sun Starfire computers that occupied many spots have been overtaken.

See also

Wikimedia Commons has media related to TOP500.

References

  1. ^ "HPL - A Portable Implementation of the High-Performance Linpack Benchmark for Distributed-Memory Computers". ICL - UTK Computer Science Department. 24 February 2016. Retrieved 22 September 2016. {{cite web}}: Unknown parameter |authors= ignored (help)
  2. ^ "AN INTERVIEW WITH JACK DONGARRA by Alan Beck, editor in chief HPCwire". Archived from the original on 28 September 2007. {{cite web}}: Unknown parameter |deadurl= ignored (|url-status= suggested) (help)
  3. ^ a b Statistics on Manufacturers and Continents
  4. ^ "The TOP25 Supercomputer Sites". Archived from the original on 23 January 2016. Retrieved 4 January 2015. {{cite web}}: Unknown parameter |deadurl= ignored (|url-status= suggested) (help)
  5. ^ "Where does Asia stand? This rising supercomputing power is reaching for real-world HPC leadership". Retrieved 4 January 2015.
  6. ^ Rpeak – This is the theoretical peak performance of the system. Measured in PFLOPS.
  7. ^ "Sublist Generator". Retrieved 4 January 2015.
  8. ^ "Top500 - List Statistics - November 2015". Top500.org. Retrieved 17 November 2015.
  9. ^ "IBM Flex System p460, POWER7 8C 3.550GHz, Infiniband QDR". TOP500 Supercomputer Sites.
  10. ^ "Microsoft Windows Azure".
  11. ^ "Magic Cube - Dawning 5000A, QC Opteron 1.9 GHz, Infiniband, Windows HPC 2008".
  12. ^ a b "November 2016 - TOP500 Supercomputer Sites".
  13. ^ a b "China Tops Supercomputer Rankings with New 93-Petaflop Machine - TOP500 Supercomputer Sites".
  14. ^ "November 2015 - TOP500 Supercomputer Sites".
  15. ^ "June 2015 - TOP500 Supercomputer Sites".
  16. ^ "November 2014 - TOP500 Supercomputer Sites".
  17. ^ "List Statistics". Retrieved 20 June 2016.
  18. ^ "Sugon".
  19. ^ Balthasar, Felix. "US Government Funds $425 million to build two new Supercomputers". News Maine. Retrieved 16 November 2014.
  20. ^ "Nuclear worries stop Intel from selling chips to Chinese supercomputers". CNN. 10 April 2015. Retrieved 17 August 2016.
  21. ^ "US nuclear fears block Intel China supercomputer update".
  22. ^ Executive Order -- Creating a National Strategic Computing Initiative (Executive order), The White House - Office of the Press Secretary, 29 July 2015
  23. ^ a b Morgan, Timothy Prickett. "Inside Japan's Future Exascale ARM Supecomputer". The Next Platform. Retrieved 13 July 2016.
  24. ^ Blue Waters Opts Out of TOP500 (article), 16 November 2012
  25. ^ Kramer, William, Top500 versus Sustained Performance – Or the Ten Problems with the TOP500 List – And What to Do About Them. 21st International Conference On Parallel Architectures And Compilation Techniques (PACT12), 19–23 September 2012, Minneapolis, MN, US
  26. ^ "Roadrunner - BladeCenter QS22/LS21 Cluster, PowerXCell 8i 3.2 Ghz / Opteron DC 1.8 GHz, Voltaire Infiniband". Retrieved 4 January 2015.
  27. ^ "Cluster Platform DL360e Gen8, Xeon E5-2450 8C 2.100GHz, Gigabit Ethernet". Retrieved 4 January 2015.
  28. ^ "Faenov - Cluster Platform SL230s Gen8, Xeon E5-2670 8C 2.600GHz, Infiniband QDR - TOP500 Supercomputer Sites".
  29. ^ "Thunder - Intel Itanium2 Tiger4 1.4GHz - Quadrics". Retrieved 4 January 2015.
  30. ^ "Columbia - SGI Altix 1.5/1.6/1.66 GHz, Voltaire Infiniband". Retrieved 4 January 2015.
  31. ^ "Japan Agency for Marine -Earth Science and Technology". Retrieved 4 January 2015.
Retrieved from "https://en.wikipedia.org/w/index.php?title=TOP500&oldid=753035234"