ASCI White: Difference between revisions

Content deleted Content added

Inline

Latest revision as of 18:55, 14 January 2024

ASCI White was a supercomputer at the Lawrence Livermore National Laboratory in California, which was briefly the fastest supercomputer in the world.^[1]

It was a computer cluster based on IBM's commercial RS/6000 SP computer. 512 nodes were interconnected for ASCI White, with each node containing sixteen 375 MHz IBM POWER3-II processors. In total, the ASCI White had 8,192 processors, 6 terabytes (TB) of memory, and 160 TB of disk storage. It was almost exclusively used for large-scale computations requiring dozens, hundreds, or thousands of processors. The computer weighed 106 tons and consumed 3 MW of electricity with a further 3 MW needed for cooling. It had a theoretical processing speed of 12.3 teraFLOPS (TFLOPS). A single modern^[when?] 4U rackmount server could match these specifications while weighing under 50 kg and consuming under 2 kW of power. The system ran IBM's AIX operating system.

ASCI White was made up of three individual systems, the 512-node White, the 28-node Ice and the 68-node Frost.

The system was built in Poughkeepsie, New York. Completed in June 2000 it was transported to specially built facilities in California and officially dedicated on August 15, 2001.^[2] Its peak performance of 12.3 TFLOPS was not achieved in the widely accepted LINPACK tests. The system cost US$110 million (equivalent to $195 million in 2023).

It was built as stage three of the Accelerated Strategic Computing Initiative (ASCI) started by the U.S. Department of Energy and the National Nuclear Security Administration to build a simulator to replace live nuclear weapons testing following the moratorium imposed by the Comprehensive Test Ban Treaty started by President George H. W. Bush in 1992 and extended by Bill Clinton in 1993.

The machine was decommissioned on July 27, 2006. ^{[citation needed]}

References[edit]

^ "No. 1 system from November 2000 to November 2001". top500.org. Retrieved May 22, 2019.
^ "The World's Fastest Computer - Meeting the Challenge of Stockpile Stewardship". Archived from the original on November 11, 2014.

Records
Preceded by ASCI Red 2.379 teraflops	World's most powerful supercomputer November 2000 – November 2001	Succeeded by NEC Earth Simulator 35.86 teraflops

This supercomputer-related article is a stub. You can help Wikipedia by expanding it.

[1] "No. 1 system from November 2000 to November 2001". top500.org. Retrieved May 22, 2019.

[2] "The World's Fastest Computer - Meeting the Challenge of Stockpile Stewardship". Archived from the original on November 11, 2014.

[1]

[2]

@@ Line 1: / Line 1: @@
+{{short description|Former supercomputer in the United States}}
 {{Use mdy dates|date=February 2014}}
-{{Refimprove|date=May 2014}}
+{{More citations needed|date=May 2014}}
 '''ASCI White''' was a [[supercomputer]] at the [[Lawrence Livermore National Laboratory]] in [[California]], which was briefly the fastest supercomputer in the world.<ref>{{cite web |title=No. 1 system from November 2000 to November 2001 |url=https://www.top500.org/resources/top-systems/asci-white-lawrence-livermore-national-laboratory/ |publisher=top500.org |accessdate=22 May 2019}}</ref>
 [[File:ASCIWhite01.jpg|thumb|Fisheye view of ASCI White]]
-It was a [[computer cluster]] based on [[IBM]]'s commercial [[IBM Scalable POWERparallel|RS/6000 SP]] computer. 512 nodes were interconnected for ASCI White, with each node containing sixteen 375{{nbsp}}MHz IBM [[POWER3|POWER3-II]] [[Microprocessor|processor]]s. In total, the ASCI White had 8,192 processors, 6{{nbsp}}[[terabyte]]s (TB) of memory, and 160{{nbsp}}TB of disk storage. It was almost exclusively used for large-scale computations requiring dozens, hundreds, or thousands of processors. The computer weighed 106{{nbsp}}tons and consumed 3{{nbsp}}MW of electricity with a further 3{{nbsp}}MW needed for cooling. It had a theoretical processing speed of 12.3{{nbsp}}[[Flops|teraFLOPS]] (TFLOPS). The system ran IBM's [[IBM AIX|AIX]] operating system.
+It was a [[computer cluster]] based on [[IBM]]'s commercial [[IBM Scalable POWERparallel|RS/6000 SP]] computer. 512 nodes were interconnected for ASCI White, with each node containing sixteen 375{{nbsp}}MHz IBM [[POWER3|POWER3-II]] [[Microprocessor|processor]]s. In total, the ASCI White had 8,192 processors, 6{{nbsp}}[[terabyte]]s (TB) of memory, and 160{{nbsp}}TB of disk storage. It was almost exclusively used for large-scale computations requiring dozens, hundreds, or thousands of processors. The computer weighed 106{{nbsp}}tons and consumed 3{{nbsp}}MW of electricity with a further 3{{nbsp}}MW needed for cooling. It had a theoretical processing speed of 12.3{{nbsp}}[[Flops|teraFLOPS]] (TFLOPS). A single modern{{When|date=November 2023}} 4U rackmount server could match these specifications while weighing under 50&nbsp;kg and consuming under 2&nbsp;kW of power. The system ran IBM's [[IBM AIX|AIX]] operating system.
 ASCI White was made up of three individual systems, the 512-node White, the 28-node Ice and the 68-node Frost.
-The system was built in [[Poughkeepsie, New York]]. Completed in June 2000 it was transported to specially built facilities in California and officially dedicated on August 15, 2001.<ref>[http://computation.llnl.gov/casc/sc2001_fliers/ASCI_White/ASCI_White01.html The World's Fastest Computer - Meeting the Challenge of Stockpile Stewardship] {{webarchive|url=https://web.archive.org/web/20141111101755/http://computation.llnl.gov/casc/sc2001_fliers/ASCI_White/ASCI_White01.html |date=November 11, 2014 }}</ref> Its peak performance of 12.3{{nbsp}}TFLOPS was not achieved in the widely accepted [[LINPACK]] tests. The system cost US$110 million.
+The system was built in [[Poughkeepsie, New York]]. Completed in June 2000 it was transported to specially built facilities in California and officially dedicated on August 15, 2001.<ref>{{Cite web|url=http://computation.llnl.gov/casc/sc2001_fliers/ASCI_White/ASCI_White01.html|archiveurl=https://web.archive.org/web/20141111101755/http://computation.llnl.gov/casc/sc2001_fliers/ASCI_White/ASCI_White01.html|url-status=dead|title=The World's Fastest Computer - Meeting the Challenge of Stockpile Stewardship|archivedate=November 11, 2014}}</ref> Its peak performance of 12.3{{nbsp}}TFLOPS was not achieved in the widely accepted [[LINPACK]] tests. The system cost US$110 million (equivalent to ${{Inflation|US|110|2000}}&nbsp;million in {{Inflation/year|US}}).
-It was built as stage three of the [[Accelerated Strategic Computing Initiative]] (ASCI) started by the U.S. [[United States Department of Energy|Department of Energy]] and the [[National Nuclear Security Administration]] to build a simulator to replace live [[nuclear weapon|WMD]] testing following the [[Comprehensive Test Ban Treaty|moratorium]] on testing started by [[President of the United States|President]] [[George H. W. Bush]] in 1992 and extended by [[Bill Clinton]] in 1993.
+It was built as stage three of the [[Accelerated Strategic Computing Initiative]] (ASCI) started by the U.S. [[United States Department of Energy|Department of Energy]] and the [[National Nuclear Security Administration]] to build a simulator to replace live [[nuclear weapon]]s testing following the moratorium imposed by the [[Comprehensive Test Ban Treaty]] started by [[President of the United States|President]] [[George H. W. Bush]] in 1992 and extended by [[Bill Clinton]] in 1993.
-The machine was decommissioned beginning July 27, 2006.
+The machine was decommissioned on July 27, 2006. {{Citation needed|date=January 2024}}
 == References ==
@@ Line 27: / Line 28: @@
 {{s-aft|after=[[Earth Simulator|NEC Earth Simulator]]<br />35.86 teraflops}}
 {{S-end}}
 {{Lawrence Livermore National Laboratory|state=autocollapse}}
 [[Category:Cluster computing]]
 [[Category:Nuclear stockpile stewardship]]
 [[Category:Lawrence Livermore National Laboratory]]
 [[Category:IBM supercomputers]]
+[[Category:64-bit computers]]