View source for IBM Blue Gene

{{Short description|Series of supercomputers by IBM}}
{{About|the supercomputer|the musician|'Blue' Gene Tyranny|the albums|Blue Gene (Gene Ammons album)|and|Blue Gene (Gene Pitney album)}}
{{Infobox Computer
|name               = IBM Blue Gene
|aka                =
|logo               =
|developer          = [[IBM]]
|type               = [[Supercomputer]] platform
|first_release_date = BG/L: {{Start date|Feb 1999}}{{br}}BG/P: June 2007{{br}}BG/Q:  Nov 2011
|discontinued       = {{End date|2015}}
|photo              = [[Image:IBM Blue Gene P supercomputer.jpg|270px]]
|caption            = A Blue Gene/P supercomputer at [[Argonne National Laboratory]]
|predecessor        = [[IBM Scalable POWERparallel|IBM RS/6000 SP]];{{br}}[[QCDOC]]
|successor          = [[Summit (supercomputer)|Summit]],  [[Sierra (supercomputer)|Sierra]]
|related            =
|processor          = BG/L: [[PowerPC 440]]{{br}}BG/P: [[PowerPC 400#PowerPC 450|PowerPC 450]]{{br}}BG/Q: [[IBM A2|PowerPC A2]]
}}

[[Image:LLNL BGL Diagram.png|thumb|upright=2.0|Hierarchy of Blue Gene processing units]]

'''Blue Gene''' was an [[IBM]] project aimed at designing supercomputers that can reach operating speeds in the [[FLOPS|petaFLOPS (PFLOPS)]] range, with relatively low power consumption.

The project created three generations of supercomputers, '''Blue Gene/L''', '''Blue Gene/P''', and '''Blue Gene/Q'''. During their deployment, Blue Gene systems often led the [[TOP500]]<ref name="Top500">{{cite web|url=https://www.top500.org/lists/2004/11/|title=November 2004 - TOP500 Supercomputer Sites|website=Top500.org|access-date=13 December 2019}}</ref> and [[Green500]]<ref name=Green500>{{cite web|url=http://www.green500.org|title=Green500 - TOP500 Supercomputer Sites|website=Green500.org|access-date=13 October 2017|archive-url=https://web.archive.org/web/20160826075608/http://www.green500.org/|archive-date=26 August 2016|url-status=dead}}</ref> rankings of the most powerful and most power-efficient supercomputers, respectively. Blue Gene systems have also consistently scored top positions in the [[Graph500]] list.<ref name=Graph500>{{cite web | url=http://www.graph500.org | title=The Graph500 List | url-status=dead | archive-url=https://web.archive.org/web/20111227021230/http://www.graph500.org/ | archive-date=2011-12-27 }}</ref> The project was awarded the 2009 [[National Medal of Technology and Innovation]].<ref>{{cite web|url=http://www.techradar.com/news/computing/obama-honours-ibm-supercomputer-636869|title=Obama honours IBM supercomputer|last=Harris|first=Mark |date=September 18, 2009|website=Techradar.com|access-date=2009-09-18}}</ref>

After Blue Gene/Q, IBM focused its supercomputer efforts on the [[OpenPOWER Foundation|OpenPower]] platform, using accelerators such as [[FPGA]]s and [[GPU]]s to address the diminishing returns of [[Moore's law]].<ref>{{cite web|url=http://www.nextplatform.com/2015/04/14/supercomputing-strategy-shifts-in-a-world-without-bluegene/|title=Supercomputing Strategy Shifts in a World Without BlueGene|date=14 April 2015|website=Nextplatform.com|access-date=13 October 2017}}</ref><ref>{{cite web|url=http://www.eetimes.com/document.asp?doc_id=1324631|title=IBM to Build DoE's Next-Gen Coral Supercomputers - EE Times|website=EETimes|access-date=13 October 2017|archive-url=https://web.archive.org/web/20170430140132/http://www.eetimes.com/document.asp?doc_id=1324631|archive-date=30 April 2017|url-status=dead}}</ref>

==History==
A video presentation of the history and technology of the Blue Gene project was given at the Supercomputing 2020 conference.<ref>{{citation|title=Supercomputing 2020 conference, Test of Time award video presentation|url=https://www.youtube.com/watch?v=jaF2gUTgME4}}</ref>

In December 1999, IBM announced a US$100 million research initiative for a five-year effort to build a massively [[parallel computer]], to be applied to the study of biomolecular phenomena such as [[protein folding]].<ref>{{cite journal | url=http://www.research.ibm.com/journal/sj/402/allen.pdf |title=Blue Gene: A Vision for Protein Science using a Petaflop Supercomputer |journal=IBM Systems Journal |volume=40 |issue=2|date=2017-10-23 }}</ref> The research and development was pursued by a large multi-disciplinary team at the [[Thomas J. Watson Research Center|IBM T. J. Watson Research Center]], initially led by [[William R. Pulleyblank]].<ref>{{citation|url=http://www.businessweek.com/stories/2001-11-06/a-talk-with-the-brain-behind-blue-gene|archive-url=https://archive.today/20141211052804/http://www.businessweek.com/stories/2001-11-06/a-talk-with-the-brain-behind-blue-gene|url-status=dead|archive-date=December 11, 2014|journal=[[BusinessWeek]]|title=A Talk with the Brain behind Blue Gene|date=November 6, 2001}}</ref>
The project had two main goals: to advance understanding of the mechanisms behind protein folding via large-scale simulation, and to explore novel ideas in massively parallel machine architecture and software. Major areas of investigation included: how to use this novel platform to effectively meet its scientific goals, how to make such massively parallel machines more usable, and how to achieve performance targets at a reasonable cost, through novel machine architectures.

The initial design for Blue Gene was based on an early version of the [[Cyclops64]] architecture, designed by [[Monty Denneau]]. In parallel, Alan Gara had started working on an extension of the [[QCDOC]] architecture into a more general-purpose supercomputer. The [[United States Department of Energy|US Department of Energy]] started funding the development of this system and it became known as Blue Gene/L (L for Light). Development of the original Blue Gene architecture continued under the name Blue Gene/C (C for Cyclops) and, later, Cyclops64.

Architecture and chip logic design for the Blue Gene systems was done at the [[Thomas J. Watson Research Center|IBM T. J. Watson Research Center]], chip design was completed and chips were manufactured by [[IBM Microelectronics]], and the systems were built at [[IBM Rochester |IBM Rochester, MN]].

In November 2004 a 16-[[Cabinet (computer)|rack]] system, with each rack holding 1,024 compute nodes, achieved first place in the [[TOP500]] list, with a [[LINPACK benchmarks]] performance of 70.72&nbsp;TFLOPS.<ref name=Top500/> It thereby overtook NEC's [[Earth Simulator]], which had held the title of the fastest computer in the world since 2002. From 2004 through 2007 the Blue Gene/L installation at LLNL<ref>{{cite web |url=http://asc.llnl.gov/computing_resources/bluegenel/ |title=BlueGene/L |access-date=2007-10-05 |url-status=dead |archive-url=https://web.archive.org/web/20110718034455/https://asc.llnl.gov/computing_resources/bluegenel/ |archive-date=2011-07-18 }}</ref> gradually expanded to 104 racks, achieving 478&nbsp;TFLOPS Linpack and 596&nbsp;TFLOPS peak. The LLNL BlueGene/L installation held the first position in the TOP500 list for 3.5 years, until in June 2008 it was overtaken by IBM's Cell-based [[IBM Roadrunner|Roadrunner]] system at [[Los Alamos National Laboratory]], which was the first system to surpass the 1 PetaFLOPS mark.

While the LLNL installation was the largest Blue Gene/L installation, many smaller installations followed. The November 2006 [[TOP500]] list showed 27 computers with the ''eServer Blue Gene Solution'' architecture. For example, three racks of Blue Gene/L were housed at the [[San Diego Supercomputer Center]].

While the [[TOP500]] measures performance on a single benchmark application, Linpack, Blue Gene/L also set records for performance on a wider set of applications. Blue Gene/L was the first supercomputer ever to run over 100&nbsp;[[FLOPS|TFLOPS]] sustained on a real-world application, namely a three-dimensional molecular dynamics code (ddcMD), simulating solidification (nucleation and growth processes) of molten metal under high pressure and temperature conditions. This achievement won the 2005 [[Gordon Bell Prize]].

In June 2006, [[National Nuclear Security Administration|NNSA]] and IBM announced that Blue Gene/L achieved 207.3&nbsp;TFLOPS on a quantum chemical application ([[Qbox]]).<ref>{{Cite web|url=http://www.hpcwire.com/hpc/701665.html|archive-url=https://web.archive.org/web/20070928004334/http://www.hpcwire.com/hpc/701665.html|url-status=dead|title=hpcwire.com|archive-date=September 28, 2007}}</ref> At Supercomputing 2006,<ref>{{cite web|url=http://sc06.supercomputing.org|title=SC06|website=sc06.supercomputing.org|access-date=13 October 2017}}</ref> Blue Gene/L was awarded the winning prize in all HPC Challenge Classes of awards.<ref>{{cite web |url=http://www.hpcchallenge.org/custom/index.html?lid=103&slid=212 |title=HPC Challenge Award Competition |access-date=2006-12-03 |url-status=dead |archive-url=https://web.archive.org/web/20061211183441/http://www.hpcchallenge.org/custom/index.html?lid=103&slid=212 |archive-date=2006-12-11 }}</ref> In 2007, a team from the [[IBM Almaden Research Center]] and the [[University of Nevada, Reno|University of Nevada]] ran an [[artificial neural network]] almost half as complex as the brain of a mouse for the equivalent of a second (the network was run at 1/10 of normal speed for 10 seconds).<ref>{{cite news|title=Mouse brain simulated on computer|url=http://news.bbc.co.uk/2/hi/technology/6600965.stm|agency=BBC News|date=April 27, 2007|archive-url=https://web.archive.org/web/20070525081051/http://news.bbc.co.uk/2/hi/technology/6600965.stm|archive-date=2007-05-25}}</ref>

===The name ===
The name Blue Gene comes from what it was originally designed to do, help biologists understand the processes of [[protein folding]] and [[genetics|gene development]].<ref>{{cite web|url=http://www-03.ibm.com/ibm/history/ibm100/us/en/icons/bluegene/|archive-url=https://web.archive.org/web/20120403015415/http://www-03.ibm.com/ibm/history/ibm100/us/en/icons/bluegene/|url-status=dead|archive-date=April 3, 2012|title=IBM100 - Blue Gene|date=7 March 2012|website=03.ibm.com|access-date=13 October 2017}}</ref> "Blue" is a traditional moniker that IBM uses for many of its products and [[IBM#Brand and reputation|the company itself]]. The original Blue Gene design was renamed "Blue Gene/C" and eventually [[Cyclops64]]. The "L" in Blue Gene/L comes from "Light" as that design's original name was "Blue Light". The "P" version was designed to be a [[Petascale computing|petascale]] design. "Q" is just the letter after "P".<ref>{{cite book|url=https://books.google.com/books?id=TTC7BQAAQBAJ&q=%22Blue+Gene%2FL%22+%22light%22&pg=PA318|title=Supercomputing: 28th International Supercomputing Conference, ISC 2013, Leipzig, Germany, June 16-20, 2013. Proceedings|first1=Julian M.|last1=Kunkel|first2=Thomas|last2=Ludwig|first3=Hans|last3=Meuer|date=12 June 2013|publisher=Springer|access-date=13 October 2017|via=Google Books|isbn=9783642387500}}</ref>

===Major features===
The Blue Gene/L supercomputer was unique in the following aspects:<ref>{{ cite journal |url=http://www.research.ibm.com/journal/rd49-23.html |journal=IBM Journal of Research and Development |title=Blue Gene |volume=49 |issue=2/3 |year=2005}}</ref>
* Trading the speed of processors for lower power consumption. Blue Gene/L used low frequency and low power embedded PowerPC cores with floating-point accelerators. While the performance of each chip was relatively low, the system could achieve better power efficiency for applications that could use large numbers of nodes.
* Dual processors per node with two working modes: co-processor mode where one processor handles computation and the other handles communication; and virtual-node mode, where both processors are available to run user code, but the processors share both the computation and the communication load.
* System-on-a-chip design. Components were embedded on a single chip for each node, with the exception of 512&nbsp;MB external DRAM.
* A large number of nodes (scalable in increments of 1024 up to at least 65,536).
* Three-dimensional [[torus interconnect]] with auxiliary networks for global communications (broadcast and reductions), I/O, and management.
* Lightweight OS per node for minimum system overhead (system noise).

===Architecture===
The Blue Gene/L architecture was an evolution of the QCDSP and [[QCDOC]] architectures. Each Blue Gene/L Compute or I/O node was a single [[Application-specific integrated circuit|ASIC]] with associated [[Dynamic random access memory|DRAM]] memory chips. The ASIC integrated two 700&nbsp;MHz [[PowerPC 440]] embedded processors, each with a double-pipeline-double-precision [[floating-point unit|Floating-Point Unit]] (FPU), a [[CPU cache|cache]] sub-system with built-in DRAM controller and the logic to support multiple communication sub-systems. The dual FPUs gave each Blue Gene/L node a theoretical peak performance of 5.6&nbsp;[[FLOPS|GFLOPS (gigaFLOPS)]]. The two CPUs were not [[Cache coherency|cache coherent]] with one another.

Compute nodes were packaged two per compute card, with 16 compute cards (thus 32 nodes) plus up to 2 I/O nodes per node board. A cabinet/rack contained 32 node boards.<ref>{{cite web|url=https://asc.llnl.gov/computing_resources/bluegenel/configuration.html|title=BlueGene/L Configuration|first=Lynn|last=Kissel|website=asc.llnl.gov|access-date=13 October 2017|archive-date=17 February 2013|archive-url=https://web.archive.org/web/20130217032440/https://asc.llnl.gov/computing_resources/bluegenel/configuration.html|url-status=dead}}</ref> By the integration of all essential sub-systems on a single chip, and the use of low-power logic, each Compute or I/O node dissipated about 17 watts (including DRAMs). The low power per node allowed aggressive packaging of up to 1024 compute nodes, plus additional I/O nodes, in a standard [[19-inch rack]], within reasonable limits on electrical power supply and air cooling. The system performance metrics, in terms of [[FLOPS per watt]], FLOPS per m<sup>2</sup> of floorspace and FLOPS per unit cost, allowed scaling up to very high performance. With so many nodes, component failures were inevitable. The system was able to electrically isolate faulty components, down to a granularity of half a rack (512 compute nodes), to allow the machine to continue to run.

Each Blue Gene/L node was attached to three parallel communications networks: a [[dimension|3D]] [[torus interconnect|toroidal network]] for peer-to-peer communication between compute nodes, a collective network for collective communication (broadcasts and reduce operations), and a global interrupt network for [[Barrier (computer science)|fast barriers]]. The I/O nodes, which run the [[Linux]] [[operating system]], provided communication to storage and external hosts via an [[Ethernet]] network. The I/O nodes handled filesystem operations on behalf of the compute nodes. A separate and private [[Ethernet]] management network provided access to any node for configuration, [[booting]] and diagnostics.

To allow multiple programs to run concurrently, a Blue Gene/L system could be partitioned into electronically isolated sets of nodes. The number of nodes in a partition had to be a positive [[integer]] power of 2, with at least 2<sup>5</sup> = 32 nodes. To run a program on Blue Gene/L, a partition of the computer was first to be reserved. The program was then loaded and run on all the nodes within the partition, and no other program could access nodes within the partition while it was in use. Upon completion, the partition nodes were released for future programs to use.

Blue Gene/L compute nodes used a minimal [[operating system]] supporting a single user program. Only a subset of [[POSIX]] calls was supported, and only one process could run at a time on a node in co-processor mode—or one process per CPU in virtual mode. Programmers needed to implement [[green threads]] in order to simulate local concurrency. Application development was usually performed in [[C (programming language)|C]], [[C++]], or [[Fortran]] using [[Message Passing Interface|MPI]] for communication. However, some scripting languages such as [[Ruby (programming language)|Ruby]]<ref>{{Cite web|title=Compute Node Ruby for Bluegene/L|website=www.ece.iastate.edu|url=http://www.ece.iastate.edu/~crb002/cnr.html|archive-url=https://web.archive.org/web/20090211071506/http://www.ece.iastate.edu:80/~crb002/cnr.html|url-status=dead|archive-date=February 11, 2009}}</ref> and [[Python (programming language)|Python]]<ref>{{cite conference |url=http://us.pycon.org/2011/home/ |title=Python for High Performance Computing |author=William Scullin |date=March 12, 2011 |location=Atlanta, GA}}</ref> have been ported to the compute nodes.

IBM published BlueMatter, the application developed to exercise Blue Gene/L, as open source.<ref>[https://github.com/IBM/BlueMatter Blue Matter source code, retrieved February 28, 2020]</ref> This serves to document how the torus and collective interfaces were used by applications, and may serve as a base for others to exercise the current generation of supercomputers.

==Blue Gene/P==
[[Image:BlueGeneP rack.jpg|right|thumbnail|A Blue Gene/P node card]]
[[Image:BlueGeneP schema.png|right|thumbnail|A schematic overview of a Blue Gene/P supercomputer]]

In June 2007, IBM unveiled '''Blue Gene/P''', the second generation of the Blue Gene series of supercomputers and designed through a collaboration that included IBM, LLNL, and [[Argonne National Laboratory]]'s [http://www.alcf.anl.gov/ Leadership Computing Facility].<ref>{{cite web |url=http://www-03.ibm.com/press/us/en/pressrelease/21791.wss |archive-url=https://web.archive.org/web/20070708225400/http://www-03.ibm.com/press/us/en/pressrelease/21791.wss |url-status=dead |archive-date=July 8, 2007 |title=IBM Triples Performance of World's Fastest, Most Energy-Efficient Supercomputer |date=2007-06-27 |access-date=2011-12-24}}</ref>

===Design===
The design of Blue Gene/P is a technology evolution from Blue Gene/L. Each Blue Gene/P Compute chip contains four [[PowerPC 400#PowerPC 450|PowerPC 450]] processor cores, running at 850&nbsp;MHz. The cores are [[cache coherence|cache coherent]] and the chip can operate as a 4-way [[Symmetric multiprocessing|symmetric multiprocessor]] (SMP). The memory subsystem on the chip consists of small private L2 caches, a central shared 8&nbsp;MB L3 cache, and dual [[DDR2 SDRAM|DDR2]] memory controllers. The chip also integrates the logic for node-to-node communication, using the same network topologies as Blue Gene/L, but at more than twice the bandwidth. A compute card contains a Blue Gene/P chip with 2 or 4&nbsp;GB DRAM, comprising a "compute node". A single compute node has a peak performance of 13.6&nbsp;GFLOPS. 32 Compute cards are plugged into an air-cooled node board. A [[19-inch rack|rack]] contains 32 node boards (thus 1024 nodes, 4096 processor cores).<ref>{{ cite journal |title=Overview of the IBM Blue Gene/P project |journal=IBM Journal of Research and Development |date=Jan 2008 |doi=10.1147/rd.521.0199 |volume=52 |pages=199–220}}</ref>
By using many small, low-power, densely packaged chips, Blue Gene/P exceeded the [[Performance per watt|power efficiency]] of other supercomputers of its generation, and at 371&nbsp;[[FLOPS per watt|MFLOPS/W]] Blue Gene/P installations ranked at or near the top of the [[Green500]] lists in 2007–2008.<ref name=Green500/>

===Installations===
The following is an incomplete list of Blue Gene/P installations. Per November 2009, the [[TOP500]] list contained 15 Blue Gene/P installations of 2-racks (2048 nodes, 8192 processor cores, 23.86&nbsp;[[FLOPS|TFLOPS]] [[Linpack]]) and larger.<ref name=Top500/>
* On November 12, 2007, the first Blue Gene/P installation, [[JUGENE]], with 16 racks (16,384 nodes, 65,536 processors) was running at [[Forschungszentrum Jülich]] in [[Germany]] with a performance of 167&nbsp;TFLOPS.<ref>{{cite news | title =Supercomputing: Jülich Amongst World Leaders Again | publisher = IDG News Service | date = 2007-11-12 | url=http://www.pressebox.de/pressemeldungen/ibm-deutschland-gmbh-4/boxid-136200.html}}</ref> When inaugurated it was the fastest supercomputer in Europe and the sixth fastest in the world. In 2009, JUGENE was upgraded to 72 racks (73,728 nodes, 294,912 processor cores) with 144 terabytes of memory and 6 petabytes of storage, and achieved a peak performance of 1 PetaFLOPS. This configuration incorporated new air-to-water heat exchangers between the racks, reducing the cooling cost substantially.<ref>{{cite web|url=http://www-03.ibm.com/press/us/en/pressrelease/26657.wss |archive-url=https://web.archive.org/web/20090212135256/http://www-03.ibm.com/press/us/en/pressrelease/26657.wss |url-status=dead |archive-date=February 12, 2009 |title=IBM Press room - 2009-02-10 New IBM Petaflop Supercomputer at German Forschungszentrum Juelich to Be Europe's Most Powerful |publisher=03.ibm.com |date=2009-02-10 |access-date=2011-03-11}}</ref> JUGENE was shut down in July 2012 and replaced by the Blue Gene/Q system [[JUQUEEN]].
* The 40-rack (40960 nodes, 163840 processor cores) "Intrepid" system at [[Argonne National Laboratory]] was ranked #3 on the June 2008 Top 500 list.<ref>{{cite web|url=http://www.mcs.anl.gov/news/detail.php?id=147|title=Argonne's Supercomputer Named World's Fastest for Open Science, Third Overall|website=Mcs.anl.gov|access-date=13 October 2017|archive-url=https://web.archive.org/web/20090208225756/http://www.mcs.anl.gov/news/detail.php?id=147|archive-date=8 February 2009|url-status=dead}}</ref> The Intrepid system is one of the major resources of the INCITE program, in which processor hours are awarded to "grand challenge" science and engineering projects in a peer-reviewed competition.
* [[Lawrence Livermore National Laboratory]] installed a 36-rack Blue Gene/P installation, "Dawn", in 2009.
* The [[King Abdullah University of Science and Technology]] ([[KAUST]]) installed a 16-rack Blue Gene/P installation, "[[Shaheen (supercomputer)|Shaheen]]", in 2009.
* In 2012, a 6-rack Blue Gene/P was installed at [[Rice University]] and will be jointly administered with the [[University of São Paulo]].<ref>{{Cite web|url=http://news.rice.edu/2012/03/30/rice-university-ibm-partner-to-bring-first-blue-gene-supercomputer-to-texas/|title=Rice University, IBM partner to bring first Blue Gene supercomputer to Texas|website=news.rice.edu|access-date=2012-04-01|archive-date=2012-04-05|archive-url=https://web.archive.org/web/20120405051308/http://news.rice.edu/2012/03/30/rice-university-ibm-partner-to-bring-first-blue-gene-supercomputer-to-texas/|url-status=dead}}</ref>
* A 2.5 rack Blue Gene/P system is the central processor for the Low Frequency Array for Radio astronomy ([[Low-Frequency Array (LOFAR)|LOFAR]]) project in the Netherlands and surrounding European countries. This application uses the streaming data capabilities of the machine.
* A 2-rack Blue Gene/P was installed in September 2008 in [[Sofia]], [[Bulgaria]], and is operated by the [[Bulgarian Academy of Sciences]] and [[Sofia University]].<ref>[http://dnes.dir.bg/2008/09/09/news3363693.html#sepultura Вече си имаме и суперкомпютър] {{Webarchive|url=https://web.archive.org/web/20091223090815/http://dnes.dir.bg/2008/09/09/news3363693.html#sepultura |date=2009-12-23 }}, Dir.bg, 9 September 2008</ref>
* In 2010, a 2-rack (8192-core) Blue Gene/P was installed at the [[University of Melbourne]] for the [[Victorian Life Sciences Computation Initiative]].<ref>{{cite web|url=http://www-03.ibm.com/press/au/en/pressrelease/29383.wss |archive-url=https://archive.today/20120716051136/http://www-03.ibm.com/press/au/en/pressrelease/29383.wss |url-status=dead |archive-date=July 16, 2012 |title=IBM Press room - 2010-02-11 IBM to Collaborate with Leading Australian Institutions to Push the Boundaries of Medical Research - Australia |publisher=03.ibm.com |date=2010-02-11 |access-date=2011-03-11}}</ref>
* In 2011, a 2-rack Blue Gene/P was installed at [[University of Canterbury]] in Christchurch, New Zealand.
* In 2012, a 2-rack Blue Gene/P was installed at [[Rutgers University]] in Piscataway, New Jersey. It was dubbed "Excalibur" as an homage to the Rutgers mascot, the Scarlet Knight.<ref>{{cite web |url=http://www.informationweek.com/hardware/supercomputers/rutgers-gets-big-data-weapon-in-ibm-supe/232700313 |title=Rutgers Gets Big Data Weapon in IBM Supercomputer - Hardware - |access-date=2013-09-07 |url-status=dead |archive-url=https://web.archive.org/web/20130306182855/http://www.informationweek.com/hardware/supercomputers/rutgers-gets-big-data-weapon-in-ibm-supe/232700313 |archive-date=2013-03-06 }}</ref>
* In 2008, a 1-rack (1024 nodes) Blue Gene/P with 180 TB of storage was installed at the [[University of Rochester]] in [[Rochester, New York]].<ref>{{cite web|title=University of Rochester and IBM Expand Partnership in Pursuit of New Frontiers in Health|url=http://www.urmc.rochester.edu/news/story/index.cfm?id=3498|publisher=University of Rochester Medical Center|archive-url=https://web.archive.org/web/20120511152144/http://www.urmc.rochester.edu/news/story/index.cfm?id=3498|archive-date=2012-05-11|date=May 11, 2012}}</ref>
* The first Blue Gene/P in the ASEAN region was installed in 2010 at the [[Universiti Brunei Darussalam|Universiti of Brunei Darussalam]]’s research centre, the [[UBD IBM Centre|UBD-IBM Centre]]. The installation has prompted research collaboration between the university and IBM research on climate modeling that will investigate the [[effects of climate change|impact of climate change]] on flood forecasting, crop yields, renewable energy and the health of rainforests in the region among others.<ref>{{cite web|title=IBM and Universiti Brunei Darussalam to Collaborate on Climate Modeling Research|url=http://www-03.ibm.com/press/us/en/pressrelease/32755.wss|archive-url=https://web.archive.org/web/20101212031355/http://www-03.ibm.com/press/us/en/pressrelease/32755.wss|url-status=dead|archive-date=December 12, 2010|publisher=IBM News Room|access-date=18 October 2012|date=2010-10-13}}</ref>
* In 2013, a 1-rack Blue Gene/P was donated to the Department of Science and Technology for weather forecasts, disaster management, precision agriculture, and health it is housed in the National Computer Center, Diliman, Quezon City, under the auspices of Philippine Genome Center (PGC) Core Facility for Bioinformatics (CFB) at UP Diliman, Quezon City.<ref>{{cite web|url=http://www.philstar.com/headlines/2014/04/15/1312838/dosts-supercomputer-scientists-now-operational|title=DOST's supercomputer for scientists now operational|first=Rainier Allan|last=Ronda|website=Philstar.com|access-date=13 October 2017}}</ref>

===Applications===
* [[Veselin Topalov]], the challenger to the [[World Chess Championship 2010|World Chess Champion]] title in 2010, confirmed in an interview that he had used a Blue Gene/P supercomputer during his preparation for the match.<ref>{{cite web|url=http://players.chessdom.com/veselin-topalov/topalov-blue-gene-p|title=Topalov training with super computer Blue Gene P|website=Players.chessdo.com|access-date=13 October 2017|archive-date=19 May 2013|archive-url=https://web.archive.org/web/20130519190040/http://players.chessdom.com/veselin-topalov/topalov-blue-gene-p|url-status=dead}}</ref>
* The Blue Gene/P computer has been used to simulate approximately one percent of a human cerebral cortex, containing 1.6 billion [[neuron]]s with approximately 9 trillion connections.<ref>Kaku, Michio. [[Physics of the Future]] (New York: Doubleday, 2011), 91.</ref>
* The [[IBM Kittyhawk]] project team has ported Linux to the compute nodes and demonstrated generic Web 2.0 workloads running at scale on a Blue Gene/P. Their paper, published in the ACM Operating Systems Review, describes a kernel driver that tunnels Ethernet over the tree network, which results in all-to-all [[TCP/IP]] connectivity.<ref>{{cite web |url=http://www.research.ibm.com/kittyhawk/ |title=Project Kittyhawk: A Global-Scale Computer|website=Research.ibm.com|access-date=13 October 2017}}</ref><ref>{{cite web |url=http://weather.ou.edu/~apw/projects/kittyhawk/kittyhawk.pdf |title=Project Kittyhawk: Building a Global-Scale Computer |first1=Jonathan |last1=Appavoo |first2=Volkmar |last2=Uhlig |first3=Amos |last3=Waterland |publisher=IBM T.J. Watson Research Center |location=Yorktown Heights, NY |access-date=2018-03-13 |archive-date=2008-10-31 |archive-url=https://wayback.archive-it.org/all/20081031010631/http://weather.ou.edu/~apw/projects/kittyhawk/kittyhawk.pdf |url-status=bot: unknown }}</ref> Running standard Linux software like [[MySQL]], their performance results on SpecJBB rank among the highest on record.{{Citation needed|date=September 2010}}
* In 2011, a Rutgers University / IBM / University of Texas team linked the [[KAUST]] [[Shaheen (supercomputer)|Shaheen]] installation together with a Blue Gene/P installation at the [[IBM Watson Research Center]] into a "federated high performance computing cloud", winning the IEEE SCALE 2011 challenge with an oil reservoir optimization application.<ref>{{cite web |url=http://news.rutgers.edu/medrel/special-content/summer-2011/rutgers-led-experts-20110706 |title=Rutgers-led Experts Assemble Globe-Spanning Supercomputer Cloud |website=News.rutgers.edu |date=2011-07-06 |access-date=2011-12-24 |url-status=dead |archive-url=https://web.archive.org/web/20111110044858/http://news.rutgers.edu/medrel/special-content/summer-2011/rutgers-led-experts-20110706 |archive-date=2011-11-10 }}</ref>

==Blue Gene/Q==
[[File:Mira - Blue Gene Q at Argonne National Laboratory - Skin.jpg|thumb|The IBM Blue Gene/Q installation ''[[Mira (supercomputer)|Mira]]'' at the [[Argonne National Laboratory]], near Chicago, Illinois]]
The third design in the Blue Gene series, '''Blue Gene/Q''', significantly expanded and enhanced on the Blue Gene/L and /P architectures.

=== Design ===
The Blue Gene/Q "compute chip" is based on the [[64-bit]] [[IBM A2]] processor core. The A2 processor core is 4-way [[simultaneous multithreading|simultaneously multithreaded]]  and was augmented with a [[SIMD]] quad-vector [[double-precision floating-point format|double-precision]] [[floating-point arithmetic|floating-point]] unit (IBM QPX). Each Blue Gene/Q compute chip contains 18 such A2 processor cores, running at 1.6&nbsp;GHz. 16 Cores are used for application computing and a 17th core is used for handling operating system assist functions such as [[interrupt]]s, [[asynchronous I/O]], [[Message Passing Interface|MPI]] pacing, and [[Reliability, availability and serviceability|RAS]]. The 18th core is a [[redundancy (engineering)|redundant]] manufacturing spare, used to increase yield. The spared-out core is disabled prior to system operation. The chip's processor cores are linked by a crossbar switch to a 32&nbsp;MB [[eDRAM]] L2 cache, operating at half core speed. The L2 cache is multi-versioned—supporting [[transactional memory]] and [[speculative execution]]—and has hardware support for [[atomic operations]].<ref name="PACT11">{{cite web
|title=Memory Speculation of the Blue Gene/Q Compute Chip
|url=http://wands.cse.lehigh.edu/IBM_BQC_PACT2011.ppt
|access-date=2011-12-23 }}</ref> L2 cache misses are handled by two built-in [[DDR3]] memory controllers running at 1.33&nbsp;GHz. The chip also integrates logic for chip-to-chip communications in a 5D [[Torus interconnect|torus]] configuration, with 2&nbsp;GB/s chip-to-chip links. The Blue Gene/Q chip is manufactured on IBM's copper SOI process at 45&nbsp;nm. It delivers a peak performance of 204.8&nbsp;GFLOPS while drawing approximately 55 watts. The chip measures 19×19&nbsp;mm (359.5&nbsp;mm²) and comprises 1.47&nbsp;billion transistors. Completing the compute node, the chip is mounted on a compute card along with 16&nbsp;GB [[DDR3]] [[DRAM]] (i.e., 1&nbsp;GB for each user processor core).<ref name="HotChips11">{{cite web
|title=The Blue Gene/Q Compute chip
|url=http://www.hotchips.org/wp-content/uploads/hc_archives/hc23/HC23.18.1-manycore/HC23.18.121.BlueGene-IBM_BQC_HC23_20110818.pdf
|access-date=2011-12-23
|archive-date=2015-04-29
|archive-url=https://web.archive.org/web/20150429070233/http://www.hotchips.org/wp-content/uploads/hc_archives/hc23/HC23.18.1-manycore/HC23.18.121.BlueGene-IBM_BQC_HC23_20110818.pdf
|url-status=dead
}}</ref>

A Q32<ref>{{cite web|url=http://www-01.ibm.com/common/ssi/rep_ca/8/897/ENUS112-028/ENUS112-028.PDF|title=IBM Blue Gene/Q supercomputer delivers petascale computing for high-performance computing applications|website=01.ibm.com|access-date=13 October 2017}}</ref> "compute drawer" contains 32 compute nodes, each water cooled.<ref name="TheRegSC10">{{cite web
|title=IBM uncloaks 20 petaflops BlueGene/Q super
|website=The Register
|date=2010-11-22
|url=https://www.theregister.co.uk/2010/11/22/ibm_blue_gene_q_super/
|access-date=2010-11-25}}</ref>
A "midplane" (crate) contains 16 Q32 compute drawers for a total of 512 compute nodes, electrically interconnected in a 5D torus configuration (4x4x4x4x2). Beyond the midplane level, all connections are optical. Racks have two midplanes, thus 32 compute drawers, for a total of 1024 compute nodes, 16,384 user cores, and 16 TB RAM.<ref name="TheRegSC10" />

Separate I/O drawers, placed at the top of a rack or in a separate rack, are air cooled and contain 8 compute cards and 8 PCIe expansion slots for [[InfiniBand]] or [[10 Gigabit Ethernet]] networking.<ref name="TheRegSC10" />

=== Performance ===
At the time of the Blue Gene/Q system announcement in November 2011,<ref>{{cite web|title=IBM announces 20-petaflops supercomputer|url=http://www.kurzweilai.net/ibm-announces-20-petaflops-supercomputer|publisher=Kurzweil|access-date=13 November 2012|date=18 November 2011|quote=IBM has announced the Blue Gene/Q supercomputer, with peak performance of 20 petaflops}}</ref> an initial 4-rack Blue Gene/Q system (4096 nodes, 65536 user processor cores) achieved #17 in the [[TOP500]] list<ref name=Top500 /> with 677.1 TeraFLOPS Linpack, outperforming the original 2007 104-rack BlueGene/L installation described above. The same 4-rack system achieved the top position in the [[Graph500]] list<ref name=Graph500/> with over 250 GTEPS (giga [[traversed edges per second]]). Blue Gene/Q systems also topped the [[Green500]] list of most energy efficient supercomputers with up to 2.1&nbsp;[[Performance per watt|GFLOPS/W]].<ref name=Green500/>

In June 2012, Blue Gene/Q installations took the top positions in all three lists: [[TOP500]],<ref name=Top500/> [[Graph500]]<ref name=Graph500/> and [[Green500]].<ref name=Green500/>

=== Installations ===
The following is an incomplete list of Blue Gene/Q installations. Per June 2012, the TOP500 list contained 20 Blue Gene/Q installations of 1/2-rack (512 nodes, 8192 processor cores, 86.35&nbsp;TFLOPS Linpack) and larger.<ref name=Top500/>  At a (size-independent) power efficiency of about 2.1 GFLOPS/W, all these systems also populated the top of the June 2012 [[Green500#Green500 List|Green 500]] list.<ref name=Green500/>
* A Blue Gene/Q system called [[Sequoia (supercomputer)|Sequoia]] was delivered to the [[Lawrence Livermore National Laboratory]] (LLNL) beginning in 2011 and was fully deployed in June 2012. It is part of the [[Advanced Simulation and Computing Program]] running nuclear simulations and advanced scientific research. It consists of 96 racks (comprising 98,304 compute nodes with 1.6 million processor cores and 1.6&nbsp;[[Petabyte|PB]] of memory) covering an area of about {{convert|3000|sqft|m2}}.<ref>{{cite web |last=Feldman |first=Michael |url=http://www.hpcwire.com/features/Lawrence-Livermore-Prepares-for-20-Petaflop-Blue-GeneQ-38948594.html |title=Lawrence Livermore Prepares for 20 Petaflop Blue Gene/Q |publisher=HPCwire |date=2009-02-03 |access-date=2011-03-11 |url-status=dead |archive-url=https://web.archive.org/web/20090212034132/http://www.hpcwire.com/features/Lawrence-Livermore-Prepares-for-20-Petaflop-Blue-GeneQ-38948594.html |archive-date=2009-02-12 }}</ref> In June 2012, the system was ranked as the world's fastest supercomputer.<ref>{{cite web |last=B Johnston |first=Donald |url=https://www.llnl.gov/news/newsreleases/2012/Jun/NR-12-06-07.html |title=NNSA's Sequoia supercomputer ranked as world's fastest |date=2012-06-18 |access-date=2012-06-23 |archive-url=https://web.archive.org/web/20140902012510/https://www.llnl.gov/news/newsreleases/2012/Jun/NR-12-06-07.html |archive-date=2014-09-02 |url-status=dead }}</ref><ref>{{Cite web|url=http://www.top500.org/lists/2012/06/press-release|archive-url=https://web.archive.org/web/20120624041004/http://www.top500.org/lists/2012/06/press-release|url-status=dead|title=TOP500 Press Release|archive-date=June 24, 2012}}</ref> at 20.1&nbsp;[[FLOPS|PFLOPS]] peak, 16.32&nbsp;[[FLOPS|PFLOPS]] sustained (Linpack), drawing up to 7.9 [[MegaWatt|megawatts]] of power.<ref name=Top500/>  In June 2013, its performance is listed at 17.17&nbsp;[[FLOPS|PFLOPS]] sustained (Linpack).<ref name=Top500/>
* A 10 PFLOPS (peak) Blue Gene/Q system called ''[[Mira (supercomputer)|Mira]]'' was installed at [[Argonne National Laboratory]] in the [http://www.alcf.anl.gov/ Argonne Leadership Computing Facility] in 2012. It consist of 48 racks (49,152 compute nodes), with 70&nbsp;[[Petabyte|PB]] of disk storage (470&nbsp;GB/s I/O bandwidth).<ref>{{cite web|url=http://www.alcf.anl.gov/articles/mira-worlds-fastest-supercomputer|title=MIRA: World's fastest supercomputer - Argonne Leadership Computing Facility|website=Alcf.anl.gov|access-date=13 October 2017}}</ref><ref>{{cite web|url=http://www.alcf.anl.gov/mira|title=Mira - Argonne Leadership Computing Facility|website=Alcf.anl.gov|access-date=13 October 2017}}</ref>
* ''JUQUEEN'' at the [[Forschungszentrum Jülich|Forschungzentrum Jülich]] is a 28-rack Blue Gene/Q system, and was from June 2013 to November 2015 the highest ranked machine in Europe in the Top500.<ref name=Top500/>
* ''Vulcan'' at  [[Lawrence Livermore National Laboratory]] (LLNL) is a 24-rack, 5&nbsp;PFLOPS (peak), Blue Gene/Q system that was commissioned in 2012 and decommissioned in 2019.<ref>{{cite web|url=https://hpc.llnl.gov/hardware/platforms/vulcan-decommissioned|website=hpc.llnl.gov|title=Vulcan—decommissioned|access-date=10 April 2019}}</ref> Vulcan served Lab-industry projects through Livermore's High Performance Computing (HPC) Innovation Center<ref>{{cite web|url=http://hpcinnovationcenter.llnl.gov/|title=HPC Innovation Center|website=hpcinnovationcenter.llnl.gov|access-date=13 October 2017}}</ref> as well as academic collaborations in support of DOE/National Nuclear Security Administration (NNSA) missions.<ref>{{cite web|url=https://www.llnl.gov/news/newsreleases/2013/Jun/NR-13-06-05.html|title=Lawrence Livermore's Vulcan brings 5 petaflops computing power to collaborations with industry and academia to advance science and technology|date=11 June 2013|website=Llnl.gov|access-date=13 October 2017|archive-date=9 December 2013|archive-url=https://web.archive.org/web/20131209231310/https://www.llnl.gov/news/newsreleases/2013/Jun/NR-13-06-05.html|url-status=dead}}</ref>
* ''Fermi'' at the [[CINECA]] Supercomputing facility, Bologna, Italy,<ref>{{cite web |url=http://www.hpc.cineca.it/content/ibm-fermi |title=Ibm-Fermi &#124; Scai |access-date=2013-05-13 |url-status=dead |archive-url=https://web.archive.org/web/20131030045547/http://www.hpc.cineca.it/content/ibm-fermi |archive-date=2013-10-30 }}</ref> is a 10-rack, 2&nbsp;PFLOPS (peak), Blue Gene/Q system.
* As part of [[DiRAC]], the [[EPCC]] hosts a 6 rack (6144-node) Blue Gene/Q system at the [[University of Edinburgh]]<ref>{{cite web|url=https://www.epcc.ed.ac.uk/facilities/dirac|website=epcc.ed.ac.uk|title=DiRAC BlueGene/Q}}</ref>
* A five rack Blue Gene/Q system with additional compute hardware called ''AMOS'' was installed at Rensselaer Polytechnic Institute in 2013.<ref>{{cite web|url=http://news.rpi.edu/content/2013/10/03/amos-among-world%E2%80%99s-fastest-and-most-powerful-supercomputers|title=Rensselaer at Petascale: AMOS Among the World's Fastest and Most Powerful Supercomputers|website=News.rpi.edu|access-date=13 October 2017}}</ref> The system was rated at 1048.6 teraflops, the most powerful supercomputer at any private university, and third most powerful supercomputer among all universities in 2014.<ref>{{cite web|url=http://news.rpi.edu/content/2014/08/04/amos-ranks-43rd-list-world%E2%80%99s-top-500-supercomputers|title=AMOS Ranks 1st Among Supercomputers at Private American Universities|author=Michael Mullaneyvar |website=News.rpi.edi|access-date=13 October 2017}}</ref>
* An 838 TFLOPS (peak) Blue Gene/Q system called ''Avoca'' was installed at the [[Victorian Life Sciences Computation Initiative]] in June, 2012.<ref>{{cite web|url=http://themelbourneengineer.eng.unimelb.edu.au/2012/02/worlds-greenest-computer-comes-to-melbourne/|title=World's greenest supercomputer comes to Melbourne - The Melbourne Engineer|date=16 February 2012|website=Themelbourneengineer.eng.unimelb.edu.au/|access-date=13 October 2017|archive-date=2 October 2017|archive-url=https://web.archive.org/web/20171002042114/http://themelbourneengineer.eng.unimelb.edu.au/2012/02/worlds-greenest-computer-comes-to-melbourne/|url-status=dead}}</ref> This system is part of a collaboration between IBM and VLSCI, with the aims of improving diagnostics, finding new drug targets, refining treatments and furthering our understanding of diseases.<ref>{{cite web|url=http://www.vlsci.org.au/|title=Melbourne Bioinformatics - For all researchers and students based in Melbourne's biomedical and bioscience research precinct|website=Melbourne Bioinformatics|access-date=13 October 2017}}</ref> The system consists of 4 racks, with 350&nbsp;TB of storage, 65,536 cores, 64&nbsp;TB RAM.<ref>{{cite web|url=http://www.vlsci.org.au/page/computer-software-configuration|title=Access to High-end Systems - Melbourne Bioinformatics|website=Vlsci.org.au|access-date=13 October 2017}}</ref>
* A 209 TFLOPS (peak) Blue Gene/Q system was installed at the [[University of Rochester]] in July, 2012.<ref>{{cite web|url=http://www.rochester.edu/news/show.php?id=4192|title=University of Rochester Inaugurates New Era of Health Care Research|website=Rochester.edu|access-date=13 October 2017}}</ref> This system is part of the [http://www.rochester.edu/provost/hscci/ Health Sciences Center for Computational Innovation] {{Webarchive|url=https://web.archive.org/web/20121019071900/http://www.rochester.edu/provost/hscci/ |date=2012-10-19 }}, which is dedicated to the application of [[high-performance computing]] to research programs in the [[healthcare science|health sciences]]. The system consists of a single rack (1,024 compute nodes) with 400 [[Terabyte|TB]] of high-performance storage.<ref name="circ.rochester.edu">{{cite web|url=http://www.circ.rochester.edu/resources.html|title=Resources - Center for Integrated Research Computing|website=Circ.rochester.edu|access-date=13 October 2017}}</ref>
* A 209 TFLOPS peak (172&nbsp;TFLOPS LINPACK) Blue Gene/Q system called ''Lemanicus'' was installed at the [[École Polytechnique Fédérale de Lausanne|EPFL]] in March 2013.<ref>{{cite web|url=http://bluegene.epfl.ch/ |title=EPFL BlueGene/L Homepage |date= |archive-url=https://web.archive.org/web/20071210070627/http://bluegene.epfl.ch/ |access-date=2021-03-10|archive-date=2007-12-10 }}</ref> This system belongs to the Center for Advanced Modeling Science CADMOS (<ref>{{cite web|url=http://www.cadmos.org|title=À propos|first=Super|last=Utilisateur|website=Cadmos.org|access-date=13 October 2017|archive-url=https://web.archive.org/web/20160110132825/http://cadmos.org/|archive-date=10 January 2016|url-status=dead}}</ref>) which is a collaboration between the three main research institutions on the shore of the [[Lake Geneva]] in the French speaking part of Switzerland : [[University of Lausanne]], [[University of Geneva]] and [[École Polytechnique Fédérale de Lausanne|EPFL]]. The system consists of a single rack (1,024 compute nodes) with 2.1&nbsp;[[Petabyte|PB]] of IBM GPFS-GSS storage.
* A half-rack Blue Gene/Q system, with about 100&nbsp;TFLOPS (peak), called ''Cumulus'' was installed at A*STAR Computational Resource Centre, Singapore, at early 2011.<ref>{{cite web|url=https://www.acrc.a-star.edu.sg/21/hardware.html|title=A*STAR Computational Resource Centre|website=Acrc.a-star.edu.sg|access-date=2016-08-24|archive-date=2016-12-20|archive-url=https://web.archive.org/web/20161220065923/https://www.acrc.a-star.edu.sg/21/hardware.html|url-status=dead}}</ref>

===Applications===
Record-breaking science applications have been run on the BG/Q, the first to cross 10 [[petaflops]] of sustained performance. The cosmology simulation framework HACC achieved almost 14 petaflops with a 3.6 trillion particle benchmark run,<ref>{{ cite arXiv | eprint=1211.4864| title=The Universe at Extreme Scale: Multi-Petaflop Sky Simulation on the BG/Q |author1=S. Habib |author2=V. Morozov |author3=H. Finkel |author4=A. Pope |author5=K. Heitmann|author5-link=Katrin Heitmann |author6=K. Kumaran |author7=T. Peterka |author8=J. Insley |author9=D. Daniel |author10=P. Fasel |author11=N. Frontiere |author12=Z. Lukic  |name-list-style=amp | class=cs.DC | year=2012 }}</ref> while the Cardioid code,<ref>{{cite web| url=http://researcher.watson.ibm.com/researcher/view_project.php?id=2992| title=Cardioid Cardiac Modeling Project| website=Researcher.watson.ibm.com| date=25 July 2016| access-date=13 October 2017| archive-date=21 May 2013| archive-url=https://web.archive.org/web/20130521180128/http://researcher.watson.ibm.com/researcher/view_project.php?id=2992| url-status=dead}}</ref><ref>{{cite web| url=https://str.llnl.gov/Sep12/streitz.html| title=Venturing into the Heart of High-Performance Computing Simulations| website=Str.llnl.gov| access-date=13 October 2017| archive-date=14 February 2013| archive-url=https://web.archive.org/web/20130214104403/https://str.llnl.gov/Sep12/streitz.html| url-status=dead}}</ref> which models the electrophysiology of the human heart, achieved nearly 12 petaflops with a near real-time simulation, both on [[IBM Sequoia|Sequoia]]. A fully compressible flow solver has also achieved 14.4&nbsp;PFLOP/s (originally 11&nbsp;PFLOP/s) on Sequoia, 72% of the machine's nominal peak performance.<ref>{{ Cite book | chapter-url=http://dl.acm.org/citation.cfm?doid=2503210.2504565 | doi = 10.1145/2503210.2504565 | series=SC '13| date=17 November 2013| pages=1–13| isbn=9781450323789| s2cid=12651650| chapter=11 PFLOP/S simulations of cloud cavitation collapse| title=Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis| last1=Rossinelli| first1=Diego| last2=Hejazialhosseini| first2=Babak| last3=Hadjidoukas| first3=Panagiotis| last4=Bekas| first4=Costas| last5=Curioni| first5=Alessandro| last6=Bertsch| first6=Adam| last7=Futral| first7=Scott| last8=Schmidt| first8=Steffen J.| last9=Adams| first9=Nikolaus A.| last10=Koumoutsakos| first10=Petros}}</ref>

== See also ==
* [[CNK operating system]]
* [[INK (operating system)|INK operating system]]
* [[Deep Blue (chess computer)]]

==References==
{{reflist}}

==External links==
* [https://www.ibm.com/history/blue-gene IBM history: IBM Blue Gene]
* [https://web.archive.org/web/20080515235646/http://community.anitaborg.org/wiki/images/9/92/GHC07-BlueGene_salapura.pdf Next generation supercomputers - Blue Gene/P overview (pdf)]

{{S-start}}
{{s-ach|rec}}
{{s-bef|before=[[Earth Simulator|NEC Earth Simulator]]<br /><small>35.86 teraflops</small>}}
{{s-ttl|title=[[TOP500|World's most powerful supercomputer]]<br /><small>Blue Gene/L</small><br /><small>70.72 - 478.20 teraflops</small>|years=November 2004&nbsp;– November 2007}}
{{s-aft|after=[[Roadrunner (supercomputer)|IBM Roadrunner]]<br /><small>1.026 petaflops</small>}}
{{s-bef|before=[[K computer|Fujitsu K computer]]<br /><small>10.51 petaflops</small>}}
{{s-ttl|title=<small>Blue Gene/Q</small><br /><small>16.32 petaflops</small>|years=June 2012&nbsp;– November 2012}}
{{s-aft|after=[[Titan (supercomputer)|Cray Titan]]<br /><small>17.59 petaflops</small>}}
{{S-end}}
{{Lawrence Livermore National Laboratory|state=autocollapse}}
{{IBM}}

[[Category:IBM supercomputer platforms|Blue Gene]]
[[Category:Petascale computers|Blue Gene]]
[[Category:Power microprocessors]]
[[Category:Parallel computing]]
[[Category:Transactional memory]]
[[Category:32-bit computers]]
[[Category:64-bit computers]]
[[Category:Lawrence Livermore National Laboratory]]