Jump to content

Fujitsu A64FX: Difference between revisions

From Wikipedia, the free encyclopedia
Content deleted Content added
Top500.org reports Rpeak for the 2.2 GHz boost mode (same as at English-language r-ccs.riken.jp page), and Rmax lower than for the 2.0 GHz "normal" mode. For AI benchmark 2.0 exaflops has been achived, higher than "Half Precision (16 bit) 1.95 Exaflops" close to the boost "Half Precision (16 bit) 2.15 Exaflops". Interesting prefix-instruction, for FMA4, which likely implies micro-ops (and trace cache?).
m Added non-breaking space to non-template file size, bitrate, and bandwidth values (via WP:JWB)
 
(39 intermediate revisions by 25 users not shown)
Line 1: Line 1:
{{For|the AMD microprocessor|Athlon 64}}
{{Use dmy dates|date=March 2020}}
{{Use dmy dates|date=March 2020}}
{{Infobox CPU
{{Infobox CPU
Line 23: Line 24:
| sock1 =
| sock1 =
| pack1 =
| pack1 =
| arch = [[ARMv8.2-A]] with SVE and [[Server Base System Architecture|SBBA level 3]]
| arch = [[ARMv8.2-A]] with SVE and [[Server Base System Architecture|SBSA level 3]]
| microarch = In-house
| microarch = In-house
| numcores = 48 per CPU<ref name="FujitsuHotChips"/> plus optional assistant cores<ref name="2019-11-13-fujitsu-pr" /><ref name="2020-06-28-fujitsu-specs" />
| numcores = 48 per CPU<ref name="FujitsuHotChips"/> plus optional assistant cores<ref name="2019-11-13-fujitsu-pr" /><ref name="2020-06-28-fujitsu-specs" />
| predecessor = [[SPARC64 V]]
}}
}}
{{For|the AMD microprocessor|Athlon 64}}


The '''A64FX''' is a [[64-bit computing|64-bit]] [[ARM architecture]] [[microprocessor]] designed by [[Fujitsu]].<ref name="FujitsuHotChips" /><ref name="2018-08-22-fujitsu-pr" /> The processor is replacing the [[SPARC64 V]] as Fujitsu's processor for [[supercomputer]] applications.<ref name="2018-08-24-NP" /> It powers the [[Fugaku (supercomputer)|Fugaku]] supercomputer, the fastest supercomputer in the world by [[TOP500]] rankings as of June 2020.<ref name="2018-08-22-fujitsu-pr" /><ref name="2018-08-24-NP" /><ref>{{Cite web|title=Outline of the Development of the Supercomputer Fugaku {{!}} RIKEN Center for Computational Science RIKEN Website|url=https://www.r-ccs.riken.jp/en/fugaku/project/outline|access-date=2020-11-18|website=www.r-ccs.riken.jp}}</ref><ref>{{Cite web|title=Supercomputer Fugaku - Supercomputer Fugaku, A64FX 48C 2.2GHz, Tofu interconnect D {{!}} TOP500|url=https://www.top500.org/system/179807/|access-date=2020-11-18|website=www.top500.org}}</ref>
The '''A64FX''' is a [[64-bit computing|64-bit]] [[ARM architecture]] [[microprocessor]] designed by [[Fujitsu]].<ref name="FujitsuHotChips" /><ref name="2018-08-22-fujitsu-pr" /> The processor is replacing the [[SPARC64 V]] as Fujitsu's processor for [[supercomputer]] applications.<ref name="2018-08-24-NP" /> It powers the [[Fugaku (supercomputer)|Fugaku]] supercomputer, ranked in the [[TOP500]] as the fastest supercomputer in the world from June 2020, until falling to second place behind [[Frontier (supercomputer)|Frontier]] in June 2022.<ref>{{Cite web |title=June 2022 {{!}} TOP500 |url=https://www.top500.org/lists/top500/2022/06/ |access-date=2023-06-23 |website=www.top500.org}}</ref><ref name="2018-08-22-fujitsu-pr" /><ref name="2018-08-24-NP" /><ref>{{Cite web|title=Outline of the Development of the Supercomputer Fugaku {{!}} RIKEN Center for Computational Science RIKEN Website|url=https://www.r-ccs.riken.jp/en/fugaku/project/outline|access-date=2020-11-18|website=www.r-ccs.riken.jp|archive-date=23 January 2021|archive-url=https://web.archive.org/web/20210123110534/https://www.r-ccs.riken.jp/en/fugaku/project/outline|url-status=dead}}</ref>


==Design==
==Design==
Fujitsu collaborated with [[Arm Holdings|ARM]] to develop the processor; it is the first processor to use the [[AArch64#ARMv8.2-A|ARMv8.2-A]] [[Scalable Vector Extension]] SIMD instruction set with 512-bit vector implementation.<ref name="2018-08-22-fujitsu-pr" /> A prototype computer using the processors was ranked as the 159th fastest computer in the world on the [[TOP500]] list in November 2019.<ref name="top-500-prototype" /> It has "Four-operand [[fused multiply–add|FMA]] with Prefix Instruction",<ref name="FujitsuHotChips" /> i.e. MOVPRFX instruction followed by [[FMA3]] ([[AArch64|ARM]], like [[RISC]] in general, is a 3-operand machine, doesn't have space for 4 operands), which get packed into a single operation in the pipeline. For the processor the designer claim ">90% execution efficiency in (D|S|H)[[general_matrix_multiply|GEMM]] and INT16/8 [[dot product]]".<ref name="FujitsuHotChips" />
Fujitsu collaborated with [[Arm Ltd.|ARM]] to develop the processor; it is the first processor to use the [[AArch64#ARMv8.2-A|ARMv8.2-A]] [[Scalable Vector Extension]] SIMD instruction set with 512-bit vector implementation.<ref name="2018-08-22-fujitsu-pr" />


It has "Four-operand [[fused multiply–add|FMA]] with Prefix Instruction",<ref name="FujitsuHotChips" /> i.e. MOVPRFX instruction followed by [[Multiply–accumulate operation#Fused multiply–add|3-operand FMA operation]] ([[AArch64|ARM]], like [[RISC]] in general, is a 3-operand machine, with no space for four operands), which get packed into a single operation in the pipeline. For the processor the designer claim ">90% execution efficiency in (D|S|H)[[general matrix multiply|GEMM]] and INT16/8 [[dot product]]".<ref name="FujitsuHotChips" />
The processor uses 32 gigabytes of [[HBM2]] memory with a bandwidth of 1&nbsp;TB per second.<ref name="2018-08-22-fujitsu-pr" /> The processor contains 16 [[PCI Express|PCI Express generation 3]] lanes<ref name="FujitsuHotChips" /> to connect to accelerators (<!--not mentioned in source, while "NVM-based File I/O accelerator" is--> hypothetical e.g. [[general-purpose computing on graphics processing units|GPUs]] and [[field-programmable gate array|FPGAs]]). The reported transistor count is about 8.8<!--8.786--> billion.<ref name="2018-08-22-fujitsu-pr" />


The processor uses 32 gigabytes of [[HBM2]] memory with a bandwidth of 1&nbsp;TB per second.<ref name="2018-08-22-fujitsu-pr" /> The processor contains 16 [[PCI Express|PCI Express generation 3]] lanes<ref name="FujitsuHotChips" /> to connect to accelerators (<!--not mentioned in source, while "NVM-based File I/O accelerator" is. +1 the PCIe lanes are meant for storage, not accelerators.-->hypothetical e.g. [[general-purpose computing on graphics processing units|GPUs]] and [[field-programmable gate array|FPGAs]]). The processor also integrates a TofuD fabric controller with 10 ports implemented as 20 lanes of high-speed 28&nbsp;Gbps to connect multiple nodes in a cluster.<ref name="FujitsuHotChips" /> The reported transistor count is about 8.8<!--8.786--> billion.<ref name="2018-08-22-fujitsu-pr" />
Each CPU has 48 cores built as a "node"{{snd}} the Fugaku node has 4 "assistant cores", however Fujitsu intends to produce lower specification machines with zero or two assistant cores.<ref name="2019-11-13-fujitsu-pr" /><ref name="2020-06-28-fujitsu-specs" /> [[Reliability, availability and serviceability]] (RAS) capabilities are claimed, i.e. ~128,400 error checkers in total.<!-- e.g slide 18: "Register ECC (INT), Parity (Others)", assuming floating-point, is that standard to not have ECC there too?-->

Each A64FX processor has four NUMA nodes, with each NUMA node having 12 compute cores, for a total of 48 cores per processor.<ref name="2020-odajima" /><ref name="2019-11-13-fujitsu-pr" /><ref name="2020-06-28-fujitsu-specs" /> Each NUMA node has its own level 2 cache, HBM2 memory, and assistant cores for non-computational purposes.<ref name="2020-odajima" />

Fujitsu intends to produce lower specification machines with reduced assistant cores.<ref name="2019-11-13-fujitsu-pr" /><ref name="2020-06-28-fujitsu-specs" /> [[Reliability, availability and serviceability]] (RAS) capabilities are claimed, i.e. ~128,400 error checkers in total.

In June 2020 the [[Fugaku (supercomputer)|Fugaku]] supercomputer using this processor reached 442 petaFLOPS and became the fastest supercomputer in the world.


==Implementations==
==Implementations==
Fujitsu designed the A64FX for the [[Fugaku (supercomputer)|Fugaku]]. As of June 2020, the Fugaku is the fastest supercomputer in the world by [[TOP500]] rankings.<ref>{{Cite web|title=Supercomputer Fugaku - Supercomputer Fugaku, A64FX 48C 2.2GHz, Tofu interconnect D {{!}} TOP500|url=https://www.top500.org/system/179807/|access-date=2020-11-18|website=www.top500.org}}</ref> Fujitsu intends to sell smaller machines with A64FX processors.<ref name="2019-11-13-fujitsu-pr" /><ref name="2020-06-28-fujitsu-specs" /> Anandtech reported in June 2020 that the cost of a PRIMEHPC FX700 server, with 2 A64FX nodes, was {{¥|4155330|link=yes}} (c. {{US$|39000|link=yes}}).<ref name="2020-06-26-anandtech" />
Fujitsu designed the A64FX for the [[Fugaku (supercomputer)|Fugaku]]. As of June and November 2020, the Fugaku is the fastest supercomputer in the world by [[TOP500]] rankings.<ref>{{Cite web|title=Supercomputer Fugaku - Supercomputer Fugaku, A64FX 48C 2.2GHz, Tofu interconnect D {{!}} TOP500|url=https://www.top500.org/system/179807/|access-date=2020-11-18|website=www.top500.org}}</ref> Fujitsu intends to sell smaller machines with A64FX processors.<ref name="2019-11-13-fujitsu-pr" /><ref name="2020-06-28-fujitsu-specs" /> Anandtech reported in June 2020 that the cost of a PRIMEHPC FX700 server, with two A64FX nodes, was {{¥|4155330|link=yes}} (c. {{US$|39000|link=yes}}).<ref name="2020-06-26-anandtech" />


[[Cray]] is developing supercomputers using the A64FX.<ref name="2019-11-13-hpcwire" /> The Isambard&nbsp;2 supercomputer is being built for a consortium in the [[United Kingdom]], led by the [[University of Bristol]] and also including the [[Met Office]], using the Fujitsu processors.<ref name="2020-02-17-UoB-pr" /><ref name="2020-03-09-nextplatform" /> It is an upgrade to the Isambard supercomputer which was built with the Marvell [[ThunderX2]], another ARM architecture microprocessor.<ref name="2020-03-09-nextplatform" />
[[Cray]] is developing supercomputers using the A64FX.<ref name="2019-11-13-hpcwire" /><ref name="2021-01-07-japan-times" /> The {{nobr|[[Isambard 2]]}} supercomputer is being built for a consortium in the [[United Kingdom]], led by the [[University of Bristol]] and also including the [[Met Office]], using the Fujitsu processors.<ref name="2020-02-17-UoB-pr" /><ref name="2020-03-09-nextplatform" /> It is an upgrade to the Isambard supercomputer which was built with the Marvell [[ThunderX2]], another ARM architecture microprocessor.<ref name="2020-03-09-nextplatform" />

[https://www.stonybrook.edu/ookami/ Ookami] is an open testbed system supported by NSF run by [[Stony Brook University]] and the [[University at Buffalo]] providing researchers access to A64FX processors.


==See also==
==See also==
Line 51: Line 60:
==References==
==References==
{{reflist|refs=
{{reflist|refs=
<ref name="2018-08-22-fujitsu-pr">{{cite web |title=Fujitsu Successfully Triples the Power Output of Gallium-Nitride Transistors - Fujitsu Global |url=https://www.fujitsu.com/global/about/resources/news/press-releases/2018/0822-02.html |website=www.fujitsu.com |publisher=Fujitsu |accessdate=8 March 2020}}</ref>
<ref name="2018-08-22-fujitsu-pr">{{cite web |title=Fujitsu Successfully Triples the Power Output of Gallium-Nitride Transistors - Fujitsu Global |url=https://www.fujitsu.com/global/about/resources/news/press-releases/2018/0822-02.html |website=www.fujitsu.com |publisher=Fujitsu |access-date=8 March 2020}}</ref>
<ref name="2018-08-24-NP">{{cite news |last1=Morgan |first1=Timothy Prickett |title=Fujitsu’s A64FX Arm Chip Waves The HPC Banner High |url=https://www.nextplatform.com/2018/08/24/fujitsus-a64fx-arm-chip-waves-the-hpc-banner-high/ |accessdate=8 March 2020 |work=The Next Platform |date=24 August 2018}}></ref>
<ref name="2018-08-24-NP">{{cite news |last1=Morgan |first1=Timothy Prickett |title=Fujitsu's A64FX Arm Chip Waves The HPC Banner High |url=https://www.nextplatform.com/2018/08/24/fujitsus-a64fx-arm-chip-waves-the-hpc-banner-high/ |access-date=8 March 2020 |work=The Next Platform |date=24 August 2018}}></ref>

<ref name="top-500-prototype">{{cite web |title=A64FX prototype - Fujitsu A64FX, Fujitsu A64FX 48C 2GHz, Tofu interconnect D {{!}} TOP500 Supercomputer Sites |url=https://www.top500.org/system/179706 |website=www.top500.org |accessdate=8 March 2020}}</ref>
<!-- Not in use
<ref name="2019-11-13-hpcwire">{{cite news |title=Cray, Fujitsu Both Bringing Fujitsu A64FX-based Supercomputers to Market in 2020 |url=https://www.hpcwire.com/2019/11/12/cray-fujitsu-both-bringing-fujitsu-a64fx-based-supercomputers-to-market-in-2020/ |accessdate=8 March 2020 |work=HPCwire |date=13 November 2019}}</ref>
<ref name="2020-02-17-UoB-pr">{{cite web |last1=Bristol |first1=University of |title=February: GW4 Isambard - News and features - University of Bristol |url=https://www.bristol.ac.uk/news/2020/february/gw4isambard-.html |website=www.bristol.ac.uk |accessdate=8 March 2020}}</ref>
<ref name="top-500-prototype">{{cite web |title=A64FX prototype - Fujitsu A64FX, Fujitsu A64FX 48C 2GHz, Tofu interconnect D {{!}} TOP500 Supercomputer Sites |url=https://www.top500.org/system/179706 |website=www.top500.org |access-date=8 March 2020}}</ref>
Not in use-->
<ref name="2020-03-09-nextplatform">{{cite news |last1=Burt |first1=Jeffrey |title=Isambard 2 Is About Driving Technology Diversity |url=https://www.nextplatform.com/2020/03/08/isambard-2-is-about-driving-technology-diversity/ |accessdate=9 March 2020 |work=The Next Platform |date=9 March 2020}}</ref>

<ref name="FujitsuHotChips">{{cite web |title=Hot Chips 30 conference; Fujitsu briefing |url=http://www.hotchips.org/hc30/2conf/2.13_Fujitsu_HC30.Fujitsu.Yoshida.rev1.2.pdf |publisher=Toshio Yoshida}}</ref>
<ref name="2020-06-28-fujitsu-specs">{{cite web |title=FUJITSU Supercomputer PRIMEHPC Specifications|url=https://www.fujitsu.com/global/products/computing/servers/supercomputer/specifications/ |website=www.fujitsu.com |accessdate=28 June 2020}}</ref>
<ref name="2019-11-13-hpcwire">{{cite news |title=Cray, Fujitsu Both Bringing Fujitsu A64FX-based Supercomputers to Market in 2020 |url=https://www.hpcwire.com/2019/11/12/cray-fujitsu-both-bringing-fujitsu-a64fx-based-supercomputers-to-market-in-2020/ |access-date=8 March 2020 |work=HPCwire |date=13 November 2019}}</ref>
<ref name="2019-11-13-fujitsu-pr">{{cite news |title=Fujitsu Launches New PRIMEHPC Supercomputers Using Fugaku Technology - Fujitsu Global |url=https://www.fujitsu.com/global/about/resources/news/press-releases/2019/1113-02.html |accessdate=28 June 2020 |work=www.fujitsu.com |date=13 November 2019}}</ref>
<ref name="2020-02-17-UoB-pr">{{cite web |last1=Bristol |first1=University of |title=February: GW4 Isambard - News and features - University of Bristol |url=https://www.bristol.ac.uk/news/2020/february/gw4isambard-.html |website=www.bristol.ac.uk |access-date=8 March 2020}}</ref>
<ref name="2020-06-26-anandtech">{{cite news |last1=Cutress |first1=Dr Ian |title=HPC Systems Special Offer: Two A64FX Nodes in a 2U for $40k |url=https://www.anandtech.com/show/15885/hpc-systems-special-offer-two-a64fx-nodes-in-a-2u-for-40k |accessdate=28 June 2020 |work=www.anandtech.com |date=26 June 2020}}</ref>
<ref name="2020-03-09-nextplatform">{{cite news |last1=Burt |first1=Jeffrey |title=Isambard 2 Is About Driving Technology Diversity |url=https://www.nextplatform.com/2020/03/08/isambard-2-is-about-driving-technology-diversity/ |access-date=9 March 2020 |work=The Next Platform |date=9 March 2020}}</ref>
<ref name="FujitsuHotChips">{{cite web |title=Hot Chips 30 conference; Fujitsu briefing |url=http://www.hotchips.org/hc30/2conf/2.13_Fujitsu_HC30.Fujitsu.Yoshida.rev1.2.pdf |archive-url=https://web.archive.org/web/20201205202434/https://hotchips.org/hc30/2conf/2.13_Fujitsu_HC30.Fujitsu.Yoshida.rev1.2.pdf |archive-date=2020-12-05 |publisher=Toshio Yoshida}}</ref>
<ref name="2020-06-28-fujitsu-specs">{{cite web |title=FUJITSU Supercomputer PRIMEHPC Specifications|url=https://www.fujitsu.com/global/products/computing/servers/supercomputer/specifications/ |website=www.fujitsu.com |access-date=28 June 2020}}</ref>
<ref name="2019-11-13-fujitsu-pr">{{cite news |title=Fujitsu Launches New PRIMEHPC Supercomputers Using Fugaku Technology - Fujitsu Global |url=https://www.fujitsu.com/global/about/resources/news/press-releases/2019/1113-02.html |access-date=28 June 2020 |work=www.fujitsu.com |date=13 November 2019}}</ref>
<ref name="2020-06-26-anandtech">{{cite news |last1=Cutress |first1=Dr Ian |title=HPC Systems Special Offer: Two A64FX Nodes in a 2U for $40k |url=https://www.anandtech.com/show/15885/hpc-systems-special-offer-two-a64fx-nodes-in-a-2u-for-40k |access-date=28 June 2020 |work=www.anandtech.com |date=26 June 2020}}</ref>
<ref name="2020-odajima">{{cite book |last1=Odajima |first1=Tetsuya |last2=Kodama |first2=Yuetsu |last3=Tsuji |first3=Miwako |last4=Matsuda |first4=Motohiko |last5=Maruyama |first5=Yutaka |last6=Sato |first6=Mitsuhisa |title=2020 IEEE International Conference on Cluster Computing (CLUSTER) |chapter=Preliminary Performance Evaluation of the Fujitsu A64FX Using HPC Applications |date=September 2020 |pages=523–530 |doi=10.1109/CLUSTER49012.2020.00075 |isbn=978-1-7281-6677-3 |s2cid=226266547 }}</ref>
<ref name="2021-01-07-japan-times">{{cite news |last1=Tsukimori |first1=Osamu |title=Japan's Fugaku supercomputer is tackling some of the world's biggest problems |url=https://www.japantimes.co.jp/news/2021/01/07/business/tech/japans-fugaku-supercomputer/ |access-date=26 January 2021 |work=The Japan Times |date=7 January 2021}}</ref>
}}
}}


{{Fujitsu}}
[[Category:ARM architecture]]
{{Application ARM-based chips}}

[[Category:Computer-related introductions in 2019]]
[[Category:ARM processors]]
[[Category:Fujitsu microprocessors]]
[[Category:Fujitsu microprocessors]]
[[Category:64-bit microprocessors]]

Latest revision as of 17:13, 23 April 2024

A64FX
General information
Launched2019
Marketed byFujitsu
Designed byFujitsu
Common manufacturer
Architecture and classification
Technology node7 nm
MicroarchitectureIn-house
Instruction setARMv8.2-A with SVE and SBSA level 3
Physical specifications
Cores
  • 48 per CPU[1] plus optional assistant cores[2][3]
History
PredecessorSPARC64 V

The A64FX is a 64-bit ARM architecture microprocessor designed by Fujitsu.[1][4] The processor is replacing the SPARC64 V as Fujitsu's processor for supercomputer applications.[5] It powers the Fugaku supercomputer, ranked in the TOP500 as the fastest supercomputer in the world from June 2020, until falling to second place behind Frontier in June 2022.[6][4][5][7]

Design[edit]

Fujitsu collaborated with ARM to develop the processor; it is the first processor to use the ARMv8.2-A Scalable Vector Extension SIMD instruction set with 512-bit vector implementation.[4]

It has "Four-operand FMA with Prefix Instruction",[1] i.e. MOVPRFX instruction followed by 3-operand FMA operation (ARM, like RISC in general, is a 3-operand machine, with no space for four operands), which get packed into a single operation in the pipeline. For the processor the designer claim ">90% execution efficiency in (D|S|H)GEMM and INT16/8 dot product".[1]

The processor uses 32 gigabytes of HBM2 memory with a bandwidth of 1 TB per second.[4] The processor contains 16 PCI Express generation 3 lanes[1] to connect to accelerators (hypothetical e.g. GPUs and FPGAs). The processor also integrates a TofuD fabric controller with 10 ports implemented as 20 lanes of high-speed 28 Gbps to connect multiple nodes in a cluster.[1] The reported transistor count is about 8.8 billion.[4]

Each A64FX processor has four NUMA nodes, with each NUMA node having 12 compute cores, for a total of 48 cores per processor.[8][2][3] Each NUMA node has its own level 2 cache, HBM2 memory, and assistant cores for non-computational purposes.[8]

Fujitsu intends to produce lower specification machines with reduced assistant cores.[2][3] Reliability, availability and serviceability (RAS) capabilities are claimed, i.e. ~128,400 error checkers in total.

In June 2020 the Fugaku supercomputer using this processor reached 442 petaFLOPS and became the fastest supercomputer in the world.

Implementations[edit]

Fujitsu designed the A64FX for the Fugaku. As of June and November 2020, the Fugaku is the fastest supercomputer in the world by TOP500 rankings.[9] Fujitsu intends to sell smaller machines with A64FX processors.[2][3] Anandtech reported in June 2020 that the cost of a PRIMEHPC FX700 server, with two A64FX nodes, was ¥4,155,330 (c. US$39,000).[10]

Cray is developing supercomputers using the A64FX.[11][12] The Isambard 2 supercomputer is being built for a consortium in the United Kingdom, led by the University of Bristol and also including the Met Office, using the Fujitsu processors.[13][14] It is an upgrade to the Isambard supercomputer which was built with the Marvell ThunderX2, another ARM architecture microprocessor.[14]

Ookami is an open testbed system supported by NSF run by Stony Brook University and the University at Buffalo providing researchers access to A64FX processors.

See also[edit]

References[edit]

  1. ^ a b c d e f "Hot Chips 30 conference; Fujitsu briefing" (PDF). Toshio Yoshida. Archived from the original (PDF) on 5 December 2020.
  2. ^ a b c d "Fujitsu Launches New PRIMEHPC Supercomputers Using Fugaku Technology - Fujitsu Global". www.fujitsu.com. 13 November 2019. Retrieved 28 June 2020.
  3. ^ a b c d "FUJITSU Supercomputer PRIMEHPC Specifications". www.fujitsu.com. Retrieved 28 June 2020.
  4. ^ a b c d e "Fujitsu Successfully Triples the Power Output of Gallium-Nitride Transistors - Fujitsu Global". www.fujitsu.com. Fujitsu. Retrieved 8 March 2020.
  5. ^ a b Morgan, Timothy Prickett (24 August 2018). "Fujitsu's A64FX Arm Chip Waves The HPC Banner High". The Next Platform. Retrieved 8 March 2020.>
  6. ^ "June 2022 | TOP500". www.top500.org. Retrieved 23 June 2023.
  7. ^ "Outline of the Development of the Supercomputer Fugaku | RIKEN Center for Computational Science RIKEN Website". www.r-ccs.riken.jp. Archived from the original on 23 January 2021. Retrieved 18 November 2020.
  8. ^ a b Odajima, Tetsuya; Kodama, Yuetsu; Tsuji, Miwako; Matsuda, Motohiko; Maruyama, Yutaka; Sato, Mitsuhisa (September 2020). "Preliminary Performance Evaluation of the Fujitsu A64FX Using HPC Applications". 2020 IEEE International Conference on Cluster Computing (CLUSTER). pp. 523–530. doi:10.1109/CLUSTER49012.2020.00075. ISBN 978-1-7281-6677-3. S2CID 226266547.
  9. ^ "Supercomputer Fugaku - Supercomputer Fugaku, A64FX 48C 2.2GHz, Tofu interconnect D | TOP500". www.top500.org. Retrieved 18 November 2020.
  10. ^ Cutress, Dr Ian (26 June 2020). "HPC Systems Special Offer: Two A64FX Nodes in a 2U for $40k". www.anandtech.com. Retrieved 28 June 2020.
  11. ^ "Cray, Fujitsu Both Bringing Fujitsu A64FX-based Supercomputers to Market in 2020". HPCwire. 13 November 2019. Retrieved 8 March 2020.
  12. ^ Tsukimori, Osamu (7 January 2021). "Japan's Fugaku supercomputer is tackling some of the world's biggest problems". The Japan Times. Retrieved 26 January 2021.
  13. ^ Bristol, University of. "February: GW4 Isambard - News and features - University of Bristol". www.bristol.ac.uk. Retrieved 8 March 2020.
  14. ^ a b Burt, Jeffrey (9 March 2020). "Isambard 2 Is About Driving Technology Diversity". The Next Platform. Retrieved 9 March 2020.
-