Wikipedia

Floating point operations per second

(Redirected fromFLOPS)
For other uses, seeFlop.
"Operations per second" redirects here. Not to be confused withInstructions per second.

Floating point operations per second(FLOPS,flopsorflop/s) is a measure ofcomputer performanceincomputing,useful in fields of scientific computations that requirefloating-pointcalculations.[1]

For such cases, it is a more accurate measure than measuringinstructions per second.[citation needed]

Floating-point arithmetic

edit
Multipliers for flops
Name Unit Value
kiloFLOPS kFLOPS 103
megaFLOPS MFLOPS 106
gigaFLOPS GFLOPS 109
teraFLOPS TFLOPS 1012
petaFLOPS PFLOPS 1015
exaFLOPS EFLOPS 1018
zettaFLOPS ZFLOPS 1021
yottaFLOPS YFLOPS 1024
ronnaFLOPS RFLOPS 1027
quettaFLOPS QFLOPS 1030

Floating-point arithmeticis needed for very large or very smallreal numbers,or computations that require a large dynamic range. Floating-point representation is similar to scientific notation, except everything is carried out in base two, rather than base ten. The encoding scheme stores the sign, theexponent(in base two for Cray andVAX,base two or ten forIEEE floating pointformats, and base 16 forIBM Floating Point Architecture) and thesignificand(number after theradix point). While several similar formats are in use, the most common isANSI/IEEE Std. 754-1985.This standard defines the format for 32-bit numbers calledsingle precision,as well as 64-bit numbers calleddouble precisionand longer numbers calledextended precision(used for intermediate results). Floating-point representations can support a much wider range of values than fixed-point, with the ability to represent very small numbers and very large numbers.[2]

Dynamic range and precision

edit

The exponentiation inherent in floating-point computation assures a much larger dynamic range – the largest and smallest numbers that can be represented – which is especially important when processing data sets where some of the data may have extremely large range of numerical values or where the range may be unpredictable. As such, floating-point processors are ideally suited for computationally intensive applications.[3]

Computational performance

edit

FLOPS andMIPSare units of measure for the numerical computing performance of a computer. Floating-point operations are typically used in fields such as scientific computational research, as well as inmachine learning.However, before the late 1980s floating-point hardware (it's possible to implement FP arithmetic in software over any integer hardware) was typically an optional feature, and computers that had it were said to be "scientific computers", or to have "scientific computation"capability. Thus the unit MIPS was useful to measure integer performance of any computer, including those without such a capability, and to account for architecture differences, similar MOPS (million operations per second) was used as early as 1970[4]as well. Note that besides integer (or fixed-point) arithmetics, examples of integer operation include data movement (A to B) or value testing (If A = B, then C). That's why MIPS as a performance benchmark is adequate when a computer is used in database queries, word processing, spreadsheets, or to run multiple virtual operating systems.[5][6]In 1974David Kuckcoined the terms flops and megaflops for the description of supercomputer performance of the day by the number of floating-point calculations they performed per second.[7]This was much better than using the prevalent MIPS to compare computers as this statistic usually had little bearing on the arithmetic capability of the machine on scientific tasks.

FLOPS by the largestsupercomputerover time

FLOPS on an HPC-system can be calculated using this equation:[8]

This can be simplified to the most common case: a computer that has exactly 1 CPU:

FLOPS can be recorded in different measures of precision, for example, theTOP500supercomputer list ranks computers by 64 bit (double-precision floating-point format) operations per second, abbreviated toFP64.[9]Similar measures are available for32-bit(FP32) and16-bit(FP16) operations.

Floating-point operations per clock cycle for various processors

edit
Floating-point operations per clock cycle per core[10]
Microarchitecture Instruction set architecture FP64 FP32 FP16
Intel CPU
Intel 80486 x87(32-bit) ? 0.128[11] ?
x87(32-bit) ? 0.5[11] ?
MMX(64-bit) ? 1[12] ?
IntelP6Pentium III SSE(64-bit) ? 2[12] ?
IntelNetBurstPentium 4(Willamette, Northwood) SSE2(64-bit) 2 4 ?
IntelP6Pentium M SSE2(64-bit) 1 2 ?
SSE3(64-bit) 2 4 ?
4 8 ?
IntelAtom(Bonnell,Saltwell,SilvermontandGoldmont) SSE3(128-bit) 2 4 ?
IntelSandy Bridge(Sandy Bridge,Ivy Bridge) AVX(256-bit) 8 16 0
AVX2&FMA(256-bit) 16 32 0
IntelXeon Phi(Knights Corner) IMCI(512-bit) 16 32 0
AVX-512&FMA(512-bit) 32 64 0
AMD CPU
AMDBobcat AMD64(64-bit) 2 4 0
4 8 0
AMDK10 SSE4/4a(128-bit) 4 8 0
AMDBulldozer[13](Piledriver,Steamroller,Excavator)
  • AVX(128-bit) (Bulldozer, Steamroller)
  • AVX2(128-bit) (Excavator)
  • FMA3(Bulldozer)[14]
  • FMA3/4(Piledriver, Excavator)
 4 8 0
AVX2&FMA(128-bit, 256-bit decoding)[18] 8 16 0
AVX2&FMA(256-bit) 16 32 0
ARM CPU
ARM Cortex-A7, A9, A15 ARMv7 1 8 0
ARM Cortex-A32, A35 ARMv8 2 8 0
ARM Cortex-A53,A55,A57,[13]A72,A73,A75 ARMv8 4 8 0
ARM Cortex-A76,A77,A78 ARMv8 8 16 0
ARM Cortex-X1 ARMv8 16 32 ?
QualcommKrait ARMv8 1 8 0
QualcommKryo(1xx - 3xx) ARMv8 2 8 0
QualcommKryo(4xx - 5xx) ARMv8 8 16 0
SamsungExynosM1 and M2 ARMv8 2 8 0
SamsungExynosM3 and M4 ARMv8 3 12 0
IBM PowerPCA2(Blue Gene/Q) ? 8 8 (as FP64) 0
Hitachi SH-4[20][21] SH-4 1 7 0
Nvidia GPU
NvidiaCurie(GeForce 6 seriesandGeForce 7 series) PTX ? 8 ?
NvidiaTesla2.0 (GeForce GTX 260–295) PTX ? 2 ?
NvidiaFermi(only GeForce GTX 465–480, 560 Ti, 570–590) PTX 1/4 (locked by driver, 1 in hardware) 2 0
NvidiaFermi(only Quadro 600–2000) PTX 1/8 2 0
NvidiaFermi(only Quadro 4000–7000, Tesla) PTX 1 2 0
NvidiaKepler(GeForce (except Titan and Titan Black), Quadro (except K6000), Tesla K10) PTX 1/12 (forGK110:locked by driver, 2/3 in hardware) 2 0
NvidiaKepler(GeForce GTX Titan and Titan Black, Quadro K6000, Tesla (except K10)) PTX 2/3 2 0
  • NvidiaMaxwell
  • NvidiaPascal(all except Quadro GP100 and Tesla P100)
 PTX 1/16 2 1/32
NvidiaPascal(only Quadro GP100 and Tesla P100) PTX 1 2 4
NvidiaVolta[22] PTX 1 2 (FP32) + 2 (INT32) 16
NvidiaTuring(only GeForce16XX) PTX 1/16 2 (FP32) + 2 (INT32) 4
NvidiaTuring(all except GeForce16XX) PTX 1/16 2 (FP32) + 2 (INT32) 16
NvidiaAmpere[23][24](only Tesla A100/A30) PTX 2 2 (FP32) + 2 (INT32) 32
NvidiaAmpere(all GeForce and Quadro, Tesla A40/A10) PTX 1/32 2 (FP32) + 0 (INT32)or1 (FP32) + 1 (INT32) 8
AMD GPU
AMDTeraScale 1(Radeon HD 4000 series) TeraScale 1 0.4 2 ?
AMDTeraScale 2(Radeon HD 5000 series) TeraScale 2 1 2 ?
AMDTeraScale 3(Radeon HD 6000 series) TeraScale 3 1 4 ?
AMDGCN(only Radeon Pro W 8100–9100) GCN 1 2 ?
AMDGCN(all except Radeon Pro W 8100–9100, Vega 10–20) GCN 1/8 2 4
AMDGCN Vega 10 GCN 1/8 2 4
AMDGCN Vega 20(only Radeon VII) GCN 1/2 (locked by driver, 1 in hardware) 2 4
AMDGCN Vega 20(only Radeon Instinct MI50 / MI60 and Radeon Pro VII) GCN 1 2 4
RDNA 1/8 2 4
AMD RDNA3 RDNA 1/8? 4 8?
AMDCDNA CDNA 1 4 (Tensor)[27] 16
AMDCDNA 2 CDNA 2 4 (Tensor) 4 (Tensor) 16
Intel GPU
Intel Xe-LP (Iris Xe MAX)[28] Xe 1/2? 2 4
Intel Xe-HPG (Arc Alchemist)[28] Xe 0 2 16
Intel Xe-HPC (Ponte Vecchio)[29] Xe 2 2 32
Qualcomm GPU
QualcommAdreno5x0 Adreno5xx 1 2 4
QualcommAdreno6x0 Adreno6xx 1 2 4
Graphcore
Graphcore Colossus GC2[30][31] ? 0 16 64
  • Graphcore Colossus GC200 Mk2[32]
  • Graphcore Bow-2000[33]
 ? 0 32 128
Supercomputer
ENIAC@ 100 kHz in 1945 0.004[34](~0.00000003 FLOPS/W)
48-bit processor @ 208kHzinCDC 1604in 1960
60-bit processor @ 10 MHz inCDC 6600in 1964 0.3 (FP60)
60-bit processor @ 10 MHz inCDC 7600in 1967 1.0 (FP60)
Cray-1@ 80 MHz in 1976 2 (700 FLOPS/W)
CDC Cyber205 @ 50 MHz in 1981

FORTRANcompiler (ANSI 77 with vector extensions)

8 16
TransputerIMS T800-20 @ 20 MHz in 1987 0.08[35]
ParallellaE16 @ 1000 MHz in 2012 2[36](5.0 GFLOPS/W)[37]
ParallellaE64 @ 800 MHz in 2012 2[38](50.0 GFLOPS/W)[37]
Microarchitecture Instruction set architecture FP64 FP32 FP16

Performance records

edit

Single computer records

edit

In June 1997,Intel'sASCI Redwas the world's first computer to achieve one teraFLOPS and beyond. Sandia director Bill Camp said that ASCI Red had the best reliability of any supercomputer ever built, and "was supercomputing's high-water mark in longevity, price, and performance".[39]

NEC'sSX-9supercomputer was the world's firstvector processorto exceed 100 gigaFLOPS per single core.

In June 2006, a new computer was announced by Japanese research instituteRIKEN,theMDGRAPE-3.The computer's performance tops out at one petaFLOPS, almost two times faster than the Blue Gene/L, but MDGRAPE-3 is not a general purpose computer, which is why it does not appear in theTop500.orglist. It has special-purposepipelinesfor simulating molecular dynamics.

By 2007,Intel Corporationunveiled the experimentalmulti-corePOLARISchip, which achieves 1 teraFLOPS at 3.13 GHz. The 80-core chip can raise this result to 2 teraFLOPS at 6.26 GHz, although the thermal dissipation at this frequency exceeds 190 watts.[40]

In June 2007, Top500.org reported the fastest computer in the world to be theIBM Blue Gene/Lsupercomputer, measuring a peak of 596 teraFLOPS.[41]TheCray XT4hit second place with 101.7 teraFLOPS.

On June 26, 2007,IBMannounced the second generation of its top supercomputer, dubbed Blue Gene/P and designed to continuously operate at speeds exceeding one petaFLOPS, faster than the Blue Gene/L. When configured to do so, it can reach speeds in excess of three petaFLOPS.[42]

On October 25, 2007,NECCorporation of Japan issued a press release announcing its SX series modelSX-9,[43]claiming it to be the world's fastest vector supercomputer. TheSX-9features the first CPU capable of a peak vector performance of 102.4 gigaFLOPS per single core.

On February 4, 2008, theNSFand theUniversity of Texas at Austinopened full scale research runs on anAMD,Sunsupercomputer namedRanger,[44] the most powerful supercomputing system in the world for open science research, which operates at sustained speed of 0.5 petaFLOPS.

On May 25, 2008, an American supercomputer built byIBM,named 'Roadrunner', reached the computing milestone of one petaFLOPS. It headed the June 2008 and November 2008TOP500list of the most powerful supercomputers (excludinggrid computers).[45][46]The computer is located at Los Alamos National Laboratory in New Mexico. The computer's name refers to the New Mexicostate bird,thegreater roadrunner(Geococcyx californianus).[47]

In June 2008, AMD released ATI Radeon HD 4800 series, which are reported to be the first GPUs to achieve one teraFLOPS. On August 12, 2008, AMD released the ATI Radeon HD 4870X2 graphics card with twoRadeon R770GPUs totaling 2.4 teraFLOPS.

In November 2008, an upgrade to the CrayJaguar supercomputerat the Department of Energy's (DOE's) Oak Ridge National Laboratory (ORNL) raised the system's computing power to a peak 1.64 petaFLOPS, making Jaguar the world's first petaFLOPS system dedicated toopen research.In early 2009 the supercomputer was named after a mythical creature,Kraken.Kraken was declared the world's fastest university-managed supercomputer and sixth fastest overall in the 2009 TOP500 list. In 2010 Kraken was upgraded and can operate faster and is more powerful.

In 2009, theCrayJaguar performed at 1.75 petaFLOPS, beating the IBM Roadrunner for the number one spot on theTOP500list.[48]

In October 2010, China unveiled theTianhe-1,a supercomputer that operates at a peak computing rate of 2.5 petaFLOPS.[49][50]

As of 2010the fastest PCprocessorreached 109 gigaFLOPS (Intel Core i7980 XE)[51]in double precision calculations.GPUsare considerably more powerful. For example,Nvidia TeslaC2050 GPU computing processors perform around 515 gigaFLOPS[52]in double precision calculations, and the AMD FireStream 9270 peaks at 240 gigaFLOPS.[53]

In November 2011, it was announced that Japan had achieved 10.51 petaFLOPS with itsK computer.[54]It has 88,128SPARC64 VIIIfxprocessorsin 864 racks, with theoretical performance of 11.28 petaFLOPS. It is named after the Japanese word "kei",which stands for 10quadrillion,[55]corresponding to the target speed of 10 petaFLOPS.

On November 15, 2011, Intel demonstrated a single x86-based processor, code-named "Knights Corner", sustaining more than a teraFLOPS on a wide range ofDGEMMoperations. Intel emphasized during the demonstration that this was a sustained teraFLOPS (not "raw teraFLOPS" used by others to get higher but less meaningful numbers), and that it was the first general purpose processor to ever cross a teraFLOPS.[56][57]

On June 18, 2012,IBM's Sequoia supercomputer system,based at the U.S. Lawrence Livermore National Laboratory (LLNL), reached 16 petaFLOPS, setting the world record and claiming first place in the latest TOP500 list.[58]

On November 12, 2012, the TOP500 list certifiedTitanas the world's fastest supercomputer per the LINPACK benchmark, at 17.59 petaFLOPS.[59][60]It was developed by Cray Inc. at theOak Ridge National Laboratoryand combines AMD Opteron processors with "Kepler" NVIDIA Tesla graphics processing unit (GPU) technologies.[61][62]

On June 10, 2013, China'sTianhe-2was ranked the world's fastest with 33.86 petaFLOPS.[63]

On June 20, 2016, China'sSunway TaihuLightwas ranked the world's fastest with 93 petaFLOPS on the LINPACK benchmark (out of 125 peak petaFLOPS). The system was installed at the National Supercomputing Center in Wuxi, and represented more performance than the next five most powerful systems on the TOP500 list did at the time combined.[64]

In June 2019,Summit,an IBM-built supercomputer now running at the Department of Energy's (DOE) Oak Ridge National Laboratory (ORNL), captured the number one spot with a performance of 148.6 petaFLOPS on High Performance Linpack (HPL), the benchmark used to rank the TOP500 list. Summit has 4,356 nodes, each one equipped with two 22-core Power9 CPUs, and six NVIDIA Tesla V100 GPUs.[65]

In June 2022, the United States'Frontieris the most powerful supercomputer on TOP500, reaching 1102 petaFlops (1.102 exaFlops) on the LINPACK benchmarks. [66]

Distributed computing records

edit

Distributed computinguses the Internet to link personal computers to achieve more FLOPS:

  • As of April 2020,theFolding@homenetwork has over 2.3 exaFLOPS of total computing power.[67][68][69][70]It is the most powerful distributed computer network, being the first ever to break 1 exaFLOPS of total computing power. This level of performance is primarily enabled by the cumulative effort of a vast array of powerfulGPUandCPUunits.[71]

Cost of computing

edit

Hardware costs

edit
Date Approximate USD per GFLOPS Platform providing the lowest cost per GFLOPS Comments
Unadjusted 2023[77]
1945 $130 trillion $2 quadrillion ENIAC:$487,000in 1945 and $7,916,000 in 2022. $487,000/0.000000385GFLOPS.First-generation (vacuum tube-based) electronic digital computer.
1961 $20 billion $204 billion A basic installation ofIBM 7030 Stretchhad a cost at the time ofUS$7.78 millioneach. TheIBM 7030 Stretchperforms one floating-point multiply every2.4microseconds.[78]Second-generation (transistor-based) computer.
1984 $20,000,000 $100,000,000 Cray X-MP/48 $15,000,000 /0.8 GFLOPS.Third-generation (integrated circuit-based) computer.
1997 $30,000 $57,000 Two 16-processorBeowulfclusters withPentium Promicroprocessors[79]
April 2000 $1,000 $2,000 Bunyip Beowulf cluster Bunyip was the first sub-US$1/MFLOPScomputing technology. It won the Gordon Bell Prize in 2000.
May 2000 $640 $1,000 KLAT2 KLAT2 was the first computing technology which scaled to large applications while staying underUS$1/MFLOPS.[80]
August 2003 $90 $100 KASY0 KASY0 was the first sub-US$100/GFLOPScomputing technology.[81]
August 2007 $50 $70 Microwulf As of August 2007, this26 GFLOPS"personal" Beowulf cluster can be built for $1256.[82]
March 2011 $1.80 $2 HPU4Science This $30,000 cluster was built using only commercially available "gamer" grade hardware.[83]
August 2012 $0.75 $1 QuadAMD Radeon 7970System A quadAMDRadeon 7970desktop computer reaching 16 TFLOPS of single-precision, 4 TFLOPS of double-precision computing performance. Total system cost was $3000; built using only commercially available hardware.[84]
June 2013 $0.22 $0.3 Sony PlayStation 4 The SonyPlayStation 4is listed as having a peak performance of1.84TFLOPS,at a price of $400[85]
November 2013 $0.16 $0.21 AMD Sempron 145&GeForce GTX 760system Built using commercially available parts, a system using one AMDSempron145 and threeNvidiaGeForce GTX 760reaches a total of6.771 TFLOPSfor a total cost ofUS$1,090.66.[86]
December 2013 $0.12 $0.16 Pentium G550&Radeon R9 290system Built using commercially available parts.IntelPentium G550and AMDRadeon R9 290tops out at4.848 TFLOPSgrand total ofUS$681.84.[87]
January 2015 $0.08 $0.1 Celeron G1830&Radeon R9 295X2system Built using commercially available parts. IntelCeleron G1830and AMDRadeon R9 295X2tops out at over11.5 TFLOPSat a grand total ofUS$902.57.[88][89]
June 2017 $0.06 $0.07 AMD Ryzen 7 1700&AMD Radeon Vega Frontier Editionsystem Built using commercially available parts. AMD Ryzen 7 1700 CPU combined with AMD Radeon Vega FE cards in CrossFire tops out at over50 TFLOPSat just underUS$3,000for the complete system.[90]
October 2017 $0.03 $0.04 Intel Celeron G3930&AMD RX Vega 64system Built using commercially available parts. ThreeAMD RX Vega 64graphics cards provide just over 75 TFLOPS half precision (38 TFLOPS SP or 2.6 TFLOPS DP when combined with the CPU) at ~$2,050 for the complete system.[91]
November 2020 $0.03 $0.03 AMD Ryzen 3600& 3×NVIDIA RTX 3080system AMD Ryzen 3600 @ 484 GFLOPS & $199.99

3× NVIDIA RTX 3080 @ 29,770 GFLOPS each & $699.99

Total system GFLOPS = 89,794 / TFLOPS= 89.2794

Total system cost incl. realistic but low cost parts; matched with other example = $2839[92]

US$/GFLOP = $0.0314

November 2020 $0.04 $0.04 PlayStation 5 The SonyPlayStation 5Digital Edition is listed as having a peak performance of 10.28 TFLOPS (20.58 TFLOPS at half precision) at a retail price of $399.[93]
November 2020 $0.04 $0.04 Xbox Series X Microsoft'sXbox Series Xis listed as having a peak performance of 12.15 TFLOPS (24.30 TFLOPS at half precision) at a retail price of $499.[94]
September 2022 $0.02 $0.02 RTX 4090 Nvidia'sRTX 4090is listed as having a peak performance of 82.6 TFLOPS (1.32 PFLOPS at 8-bit precision) at a retail price of $1599.[95]
May 2023 $0.01 $0.01 Radeon RX 7600 AMD'sRX 7600is listed as having a peak performance of 21.5 TFLOPS at a retail price of $269.[96]


See also

edit

References

edit
  1. ^"Understand measures of supercomputer performance and storage system capacity".kb.iu.edu.RetrievedMarch 23,2024.
  2. ^Floating PointRetrieved on December 25, 2009.
  3. ^Summary: Fixed-point (integer) vs floating-pointArchivedDecember 31, 2009, at theWayback MachineRetrieved on December 25, 2009.
  4. ^NASA Technical Note.National Aeronautics and Space Administration. 1970.
  5. ^Fixed versus floating point.Retrieved on December 25, 2009.
  6. ^Data manipulation and math calculation.Retrieved on December 25, 2009.
  7. ^Kuck, D. J. (1974).Computer System Capacity Fundamentals.U.S. Department of Commerce, National Bureau of Standards.
  8. ^""Nodes, Sockets, Cores and FLOPS, Oh, My" by Dr. Mark R. Fernandez, Ph.D. "Archived fromthe originalon February 13, 2019.RetrievedFebruary 12,2019.
  9. ^"FREQUENTLY ASKED QUESTIONS".top500.org.RetrievedJune 23,2020.
  10. ^"Floating-Point Operations Per Second (FLOPS)".
  11. ^ab"home.iae.nl".
  12. ^ab"Computing Power throughout History".alternatewars.com.RetrievedFebruary 13,2021.
  13. ^abcdeDolbeau, Romain (2017). "Theoretical Peak FLOPS per instruction set: a tutorial".Journal of Supercomputing.74(3): 1341–1377.doi:10.1007/s11227-017-2177-5.S2CID3540951.
  14. ^"New instructions support for Bulldozer (FMA3) and Piledriver (FMA3+4 and CVT, BMI, TBM)"(PDF).
  15. ^"Agner's CPU blog - Test results for AMD Ryzen".
  16. ^https://arstechnica.com/gadgets/2017/03/amds-moment-of-zen-finally-an-architecture-that-can-compete/2/"each core now has a pair of 128-bit FMA units of its own"
  17. ^Mike Clark (August 23, 2016).A New x86 Core Architecture for the Next Generation of Computing(PDF).HotChips 28. AMD. Archived fromthe original(PDF)on July 31, 2020.RetrievedOctober 8,2017.page 7
  18. ^"The microarchitecture of Intel and AMD CPUs"(PDF).
  19. ^"AMD CEO Lisa Su's COMPUTEX 2019 Keynote".youtube.com.May 27, 2019.Archivedfrom the original on December 11, 2021.
  20. ^"Entertainment Systems and High-Performance Processor SH-4"(PDF).Hitachi Review.48(2).Hitachi:58–63. 1999.RetrievedJune 21,2019.
  21. ^"SH-4 Next-Generation DSP Architecture for VoIP"(PDF).Hitachi.2000.RetrievedJune 21,2019.
  22. ^"Inside Volta: The World's Most Advanced Data Center GPU".May 10, 2017.
  23. ^"NVIDIA Ampere Architecture In-Depth".May 14, 2020.
  24. ^"NVIDIA A100 GPUs Power the Modern Data Center".NVIDIA.
  25. ^Schilling, Andreas (June 10, 2019)."Die RDNA-Architektur - Seite 2".Hardwareluxx.
  26. ^"AMD Radeon RX 5700 XT Specs".TechPowerUp.
  27. ^"AMD Instinct MI100 Accelerator".
  28. ^ab"Introduction to the Xe-HPG Architecture".
  29. ^"Intel Data Center GPU Max".November 9, 2022.
  30. ^"250 TFLOPs/s for two chips with FP16 mixed precision".youtube.com.October 26, 2018.
  31. ^Archived atGhostarchiveand theWayback Machine:"Estimation via power consumption that FP32 is 1/4 of FP16 and that clock frequency is below 1.5GHz".youtube.com.October 25, 2017.
  32. ^Archived atGhostarchiveand theWayback Machine:"Introducing Graphcore's Mk2 IPU systems".youtube.com.July 15, 2020.
  33. ^"Bow-2000 IPU-Machine".docs.graphcore.ai/.
  34. ^ENIAC]] @ 100 kHz with 385 Flops"Computers of Yore".clear.rice.edu.RetrievedFebruary 26,2021.
  35. ^"IMS T800 Architecture".transputer.net.RetrievedDecember 28,2023.
  36. ^Epiphany-III 16-core 65nm Microprocessor (E16G301)//admin(August 19, 2012)
  37. ^abFeldman, Michael (August 22, 2012)."Adapteva Unveils 64-Core Chip".HPCWire.RetrievedSeptember 3,2014.
  38. ^Epiphany-IV 64-core 28nm Microprocessor (E64G401)//admin(August 19, 2012)
  39. ^"Sandia's ASCI Red, world's first teraflop supercomputer, is decommissioned"(PDF).Archived fromthe original(PDF)on November 5, 2010.RetrievedNovember 17,2011.
  40. ^Richard Swinburne (April 30, 2007)."The Arrival of TeraFLOP Computing".bit-tech.net.RetrievedFebruary 9,2012.
  41. ^"29th TOP500 List of World's Fastest Supercomputers Released".Top500.org.June 23, 2007. Archived fromthe originalon May 9, 2008.RetrievedJuly 8,2008.
  42. ^"June 2008".TOP500.RetrievedJuly 8,2008.
  43. ^"NEC Launches World's Fastest Vector Supercomputer, SX-9".NEC. October 25, 2007.RetrievedJuly 8,2008.
  44. ^"University of Texas at Austin, Texas Advanced Computing Center".Archived fromthe originalon August 1, 2009.RetrievedSeptember 13,2010.Any researcher at a U.S. institution can submit a proposal to request an allocation of cycles on the system.
  45. ^Sharon Gaudin (June 9, 2008)."IBM's Roadrunner smashes 4-minute mile of supercomputing".Computerworld. Archived fromthe originalon December 24, 2008.RetrievedJune 10,2008.
  46. ^"Austin ISC08".Top500.org. November 14, 2008. Archived fromthe originalon February 22, 2012.RetrievedFebruary 9,2012.
  47. ^Fildes, Jonathan (June 9, 2008)."Supercomputer sets petaflop pace".BBC News.RetrievedJuly 8,2008.
  48. ^Greenberg, Andy (November 16, 2009)."Cray Dethrones IBM in Supercomputing".Forbes.
  49. ^"China claims supercomputer crown".BBC News. October 28, 2010.
  50. ^Dillow, Clay (October 28, 2010)."China Unveils 2507 Petaflop Supercomputer, the World's Fastest".Popsci.com.RetrievedFebruary 9,2012.
  51. ^"Intel's Core i7-980X Extreme Edition – Ready for Sick Scores?: Mathematics: Sandra Arithmetic, Crypto, Microsoft Excel".Techgage.March 10, 2010.RetrievedFebruary 9,2012.
  52. ^"NVIDIA Tesla Personal Supercomputer".Nvidia.com.RetrievedFebruary 9,2012.
  53. ^"AMD FireStream 9270 GPU Compute Accelerator".Amd.com.RetrievedFebruary 9,2012.
  54. ^"'K computer' Achieves Goal of 10 Petaflops ".Fujitsu.com.RetrievedFebruary 9,2012.
  55. ^SeeJapanese numbers
  56. ^"Intel's Knights Corner: 50+ Core 22nm Co-processor".November 16, 2011.RetrievedNovember 16,2011.
  57. ^"Intel unveils 1 TFLOP/s Knight's Corner".RetrievedNovember 16,2011.
  58. ^Clark, Don (June 18, 2012)."IBM Computer Sets Speed Record".The Wall Street Journal.RetrievedJune 18,2012.
  59. ^"US Titan supercomputer clocked as world's fastest".BBC. November 12, 2012.RetrievedFebruary 28,2013.
  60. ^"Oak Ridge Claims No. 1 Position on Latest TOP500 List with Titan | TOP500 Supercomputer Sites".Top500.org. November 12, 2012.RetrievedFebruary 28,2013.
  61. ^Montalbano, Elizabeth (October 11, 2011)."Oak Ridge Labs Builds Fastest Supercomputer".Informationweek.RetrievedFebruary 9,2012.
  62. ^Tibken, Shara (October 29, 2012)."Titan supercomputer debuts for open scientific research | Cutting Edge".News.CNet.com.RetrievedFebruary 28,2013.
  63. ^"Chinese Supercomputer Is Now The World's Fastest – By A Lot".Forbes Magazine.June 17, 2013.RetrievedJune 17,2013.
  64. ^Feldman, Michael."China Races Ahead in TOP500 Supercomputer List, Ending US Supremacy".Top500.org.RetrievedDecember 31,2016.
  65. ^"June 2018".Top500.org.RetrievedJuly 17,2018.
  66. ^https://en.wikipedia.org/wiki/TOP500[bare URL]
  67. ^"Folding@Home Active CPUs & GPUs by OS".foldingathome.org.RetrievedApril 8,2020.
  68. ^Folding@home (March 25, 2020)."Thanks to our AMAZING community, we've crossed the exaFLOP barrier! That's over a 1,000,000,000,000,000,000 operations per second, making us ~10x faster than the IBM Summit!pic.twitter.com/mPMnb4xdH3".@foldingathome.RetrievedApril 4,2020.
  69. ^"Folding@Home Crushes Exascale Barrier, Now Faster Than Dozens of Supercomputers - ExtremeTech".extremetech.com.RetrievedApril 4,2020.
  70. ^"Folding@Home exceeds 1.5 ExaFLOPS in the battle against Covid-19".TechSpot.March 26, 2020.RetrievedApril 4,2020.
  71. ^"Sony Computer Entertainment's Support for Folding@home Project on PlayStation™3 Receives This Year's" Good Design Gold Award ""(Press release). Sony Computer Entertainment Inc. November 6, 2008. Archived fromthe originalon January 31, 2009.RetrievedDecember 11,2008.
  72. ^"BOINC Computing Power".BOINC.RetrievedDecember 28,2020.
  73. ^"SETI@Home Credit overview".BOINC.RetrievedJune 15,2018.
  74. ^"Einstein@Home Credit overview".BOINC.RetrievedJune 15,2018.
  75. ^"MilkyWay@Home Credit overview".BOINC.RetrievedJune 15,2018.
  76. ^"Internet PrimeNet Server Distributed Computing Technology for the Great Internet Mersenne Prime Search".GIMPS.RetrievedJune 15,2018.
  77. ^1634–1699:McCusker, J. J.(1997).How Much Is That in Real Money? A Historical Price Index for Use as a Deflator of Money Values in the Economy of the United States: Addenda et Corrigenda(PDF).American Antiquarian Society.1700–1799:McCusker, J. J.(1992).How Much Is That in Real Money? A Historical Price Index for Use as a Deflator of Money Values in the Economy of the United States(PDF).American Antiquarian Society.1800–present:Federal Reserve Bank of Minneapolis."Consumer Price Index (estimate) 1800–".RetrievedFebruary 29,2024.
  78. ^"The IBM 7030 (STRETCH)".Norman Hardy.RetrievedFebruary 24,2017.
  79. ^"Loki and Hyglac".Loki-www.lanl.gov. July 13, 1997. Archived fromthe originalon July 21, 2011.RetrievedFebruary 9,2012.
  80. ^"Kentucky Linux Athlon Testbed 2 (KLAT2)".The Aggregate.RetrievedFebruary 9,2012.
  81. ^"KASY0".The Aggregate.August 22, 2003.RetrievedFebruary 9,2012.
  82. ^"Microwulf: A Personal, Portable Beowulf Cluster".Archived fromthe originalon September 12, 2007.RetrievedFebruary 9,2012.
  83. ^Adam Stevenson, Yann Le Du, and Mariem El Afrit. "High-performance computing on gamer PCs."Ars Technica.March 31, 2011.
  84. ^Tom Logan (January 9, 2012)."HD7970 Quadfire Eyefinity Review".OC3D.net.
  85. ^"Sony Sparks Price War With PS4 Priced at $399."CNBC.June 11, 2013.
  86. ^"FreezePage".Archived fromthe originalon November 16, 2013.RetrievedMay 9,2020.
  87. ^"FreezePage".Archived fromthe originalon December 19, 2013.RetrievedMay 9,2020.
  88. ^"FreezePage".Archived fromthe originalon January 10, 2015.RetrievedMay 9,2020.
  89. ^"Radeon R9 295X2 8 GB Review: Project Hydra Gets Liquid Cooling".April 8, 2014.
  90. ^Perez, Carol E. (July 13, 2017)."Building a 50 Teraflops AMD Vega Deep Learning Box for Under $3K".Intuition Machine.RetrievedJuly 26,2017.
  91. ^"lowest_$/fp16 - mattebaughman's Saved Part List - Celeron G3930 2.9GHz Dual-Core, Radeon RX VEGA 64 8GB (3-Way CrossFire), XON-350_BK ATX Mid Tower".pcpartpicker.com.RetrievedSeptember 13,2017.
  92. ^"System Builder".pcpartpicker.com.RetrievedDecember 7,2020.
  93. ^"AMD Playstation 5 GPU Specs".techpowerup.com.RetrievedMay 12,2021.
  94. ^"Xbox Series X | Xbox".xbox.com.RetrievedSeptember 21,2021.
  95. ^"Nvidia Announces RTX 4090 Coming October 12, RTX 4080 Later".tomshardware.com.September 20, 2022.RetrievedSeptember 20,2022.
  96. ^"AMD Radeon RX 7600 Review: Incremental Upgrades".tomshardware.com.May 24, 2023.RetrievedMay 24,2023.
Retrieved from "https://en.wikipedia.org/w/index.php?title=Floating_point_operations_per_second&oldid=1249366560"