Talk:Supercomputer/Archive 2

While Cray was most important there were others.
I worked at Texas Instruments during the 1970's and we had a super computer the TI Advanced Research Computer. As I recall only 3 were made. One belonged to NOAA and was used for weather. One remained at TI in Austin and was used primarily for oil field seismic processing. This computer was implemented in Emmitter Coupled Logic, and could be driven fast enough to melt wires. It also was among the first to feature pipelining. I gather that the modern supercomputers use a MOS technology where the speed is bound by how small the chips can be made.

In addition to paralling, pipelining shouild be mentioned in the history segment. The underlying chip technology has changed over time and this too should be mentioned.EdEveridge (talk) 23:57, 27 November 2012 (UTC)

Full weather modelling citation
I was interested in the full weather modelling thing so was checking the citation but can't find where it supports the statement. It briefly mentions weather modelling but mostly seems to talk about climate modelling including the subset of atmospheric modelling. In particular and perhaps not surprisingly, it doesn't mention anything about a 2 week time period. In fact the modelling period it mentions is for 10000 years (page 394 & table III, page 396 & page 399) which isn't surprising since it's referring to climate modelling not weather.

I admit I only glanced thorough the paper a few times and didn't read it that carefully so I may have missed something (although I did also search for 'day' and 'week'). Particularly since my understanding of any sort of modelling, computer science or engineering is limited. But I'm not convinced the source actually supports our statement.

The later source does mention the 14 day weather modelling thing but it's from Intel and doesn't say where they got the idea.

To put it a different way from the sources I read, I don't see that it says "Erik P. DeBenedictis of Sandia National Laboratories theorizes that a zettaflop (1021, one sextillion FLOPS) computer is required to accomplish full weather modeling, which could cover a two-week time span accurately."

I also searched online but all I found were sources citing or copying us.

I had a brief look at the history and it looks to me like the statement is more or less the same as when it was added [//en.wikipedia.org/w/index.php?title=Supercomputer&diff=259192175&oldid=257869485].

Nil Einne (talk) 10:39, 5 January 2014 (UTC)

External links modified
Hello fellow Wikipedians,

I have just added archive links to 1 one external link on Supercomputer. Please take a moment to review my edit. If necessary, add after the link to keep me from modifying it. Alternatively, you can add to keep me off the page altogether. I made the following changes:
 * Attempted to fix sourcing for http://www.cosic.esat.kuleuven.ac.be/des/

When you have finished reviewing my changes, please set the checked parameter below to true to let others know.

Cheers.—cyberbot II  Talk to my owner :Online 02:25, 6 February 2016 (UTC)

New fastest supercomputer
I'm sure that the current record of Tianhe-2 being the fastest supercomputer in the world is outdated. The Top500 list just said that a new supercomputer, the Sunway TaihuLight, is now the fastest, with a speed of 93 petaflops. The link can be seen here: http://www.top500.org/news/china-tops-supercomputer-rankings-with-new-93-petaflop-machine/

Should I go ahead and make the necessary changes? — Preceding unsigned comment added by PigeonGuru (talk • contribs) 09:26, 20 June 2016 (UTC)


 * Yes, definitely. I've also made a start on an article about it, at Sunway TaihuLight. -- The Anome (talk) 10:30, 20 June 2016 (UTC)

HPC is a research agenda
I find it crazy that HPC redirects to this article (Supercomputer). Supercomputer is a thing (e.g. telescope), HPC is a research agenda (e.g. cosmology). &mdash; MaxEnt 01:44, 5 July 2016 (UTC)

I completely agree. HPC is absolutely it's own topic, and one that offers more specificity than one article will allow. — Preceding unsigned comment added by 165.225.34.101 (talk) 12:16, 9 October 2017 (UTC)

Power/Consupsion?
in the article it says "For example, Tianhe-1A consumes 4.04 megawatts (MW) of electricity." this makes no sense because there is no unit of time, so it it irrelevant. if it said 4.04 MW an hour, that would make more sense. will someone please look over this? Zjjpp iscool talk 03:08, 13 May 2017 (UTC)
 * The article is correct as stated. The Watt is a unit of power: that is, the rate at which electricity is consumed. the Watt-hour is a unit of energy. One watt-hour is the energy consumed at a rate of one Watt in one hour.  Look at your electric bill: you are charged based on Watt hours or more likely, kilowatt hours (kWh). Tianhe-1A consumes 4.04 MWh of energy for every hour that it operates at the full 4.04 MW power level. -Arch dude (talk) 06:00, 13 May 2017 (UTC)

Untitled
'''Most data on China is fake and planted as a part of the global game. China is in reality extremely backward. It is for this reason the Communist Chinese are coming to the USA, Europe, etc. while few people ever go to China. Ignore all planted data on China. ''' — Preceding unsigned comment added by 65.88.88.203 (talk) 00:16, 22 June 2017 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified 3 external links on Supercomputer. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive https://web.archive.org/web/20120330182549/http://sss.lanl.gov/pubs/031001-acmq.pdf to http://sss.lanl.gov/pubs/031001-acmq.pdf
 * Added archive https://web.archive.org/web/20120813212211/http://www.hpcwire.com/hpcwire/2010-07-02/ibm_hot_water-cooled_supercomputer_goes_live_at_eth_zurich.html to http://www.hpcwire.com/hpcwire/2010-07-02/ibm_hot_water-cooled_supercomputer_goes_live_at_eth_zurich.html
 * Added tag to http://www.green500.org/lists/2011/06/top/list.php
 * Added archive https://web.archive.org/web/20150118222233/http://www.pnl.gov/computing/resources/esdc/1_Cooling.stm to http://www.pnl.gov/computing/resources/esdc/1_Cooling.stm

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

Cheers.— InternetArchiveBot  (Report bug) 21:25, 26 July 2017 (UTC)

ALMA Correlator
Why is there a picture of the ALMA Correlator? In particular, why is there not a single mention of this in the article? The ALMA Correlator is based on on FPGAs, not CPUs or GPUs. It is a single-purpose machine and in my opinion not a real computer. You can't run arbitrary programs/computations on it. As much as it is an impressive peace of engineering, I think it is misplaced in this article. --184.189.234.199 (talk) 05:20, 14 September 2017 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified 5 external links on Supercomputer. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive https://web.archive.org/web/20120728105352/http://www.computer50.org/kgill/atlas/atlas.html to http://www.computer50.org/kgill/atlas/atlas.html
 * Added archive https://web.archive.org/web/20081217131938/http://www.hpcwire.com/topic/processors/IBM_Roadrunner_Takes_the_Gold_in_the_Petaflop_Race.html to http://www.hpcwire.com/topic/processors/IBM_Roadrunner_Takes_the_Gold_in_the_Petaflop_Race.html
 * Added archive https://web.archive.org/web/20120305234455/http://www.top500.org/overtime/list/32/os to http://www.top500.org/overtime/list/32/os
 * Added archive https://web.archive.org/web/20100919090657/http://boincstats.com/stats/project_graph.php?pr=bo to http://www.boincstats.com/stats/project_graph.php?pr=bo
 * Added archive https://web.archive.org/web/20131208075357/http://www.h-online.com/newsticker/news/item/IDF-Intel-says-Moore-s-Law-holds-until-2029-734779.html to http://www.h-online.com/newsticker/news/item/IDF-Intel-says-Moore-s-Law-holds-until-2029-734779.html

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

Cheers.— InternetArchiveBot  (Report bug) 21:09, 2 December 2017 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified one external link on Supercomputer. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive https://web.archive.org/web/20150520034755/http://www.intelfreepress.com/news/green-supercomputer-crunches-big-data-in-iceland/39/ to http://www.intelfreepress.com/news/green-supercomputer-crunches-big-data-in-iceland/39/

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

Cheers.— InternetArchiveBot  (Report bug) 01:34, 23 December 2017 (UTC)

2018 exascale computing predictions.
Under the section "Research and development trends" there's current a piece of text: "Given the current speed of progress, industry experts estimate that supercomputers will reach 1 EFLOPS (1018, 1,000 PFLOPS or one quintillion FLOPS) by 2018." Which, well, obviously hasn't happened. So should this section be edited or left as a historical note? Bryce Stansfield (talk) 03:19, 10 March 2018 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified 5 external links on Supercomputer. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive https://web.archive.org/web/20120728105352/http://www.computer50.org/kgill/atlas/atlas.html to http://www.computer50.org/kgill/atlas/atlas.html
 * Added archive https://web.archive.org/web/20081217131938/http://www.hpcwire.com/topic/processors/IBM_Roadrunner_Takes_the_Gold_in_the_Petaflop_Race.html to http://www.hpcwire.com/topic/processors/IBM_Roadrunner_Takes_the_Gold_in_the_Petaflop_Race.html
 * Added archive https://web.archive.org/web/20120305234455/http://www.top500.org/overtime/list/32/os to http://www.top500.org/overtime/list/32/os
 * Added archive https://web.archive.org/web/20100919090657/http://boincstats.com/stats/project_graph.php?pr=bo to http://www.boincstats.com/stats/project_graph.php?pr=bo
 * Added archive https://web.archive.org/web/20131208075357/http://www.h-online.com/newsticker/news/item/IDF-Intel-says-Moore-s-Law-holds-until-2029-734779.html to http://www.h-online.com/newsticker/news/item/IDF-Intel-says-Moore-s-Law-holds-until-2029-734779.html

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

Cheers.— InternetArchiveBot  (Report bug) 21:09, 2 December 2017 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified one external link on Supercomputer. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive https://web.archive.org/web/20150520034755/http://www.intelfreepress.com/news/green-supercomputer-crunches-big-data-in-iceland/39/ to http://www.intelfreepress.com/news/green-supercomputer-crunches-big-data-in-iceland/39/

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

Cheers.— InternetArchiveBot  (Report bug) 01:34, 23 December 2017 (UTC)

2018 exascale computing predictions.
Under the section "Research and development trends" there's current a piece of text: "Given the current speed of progress, industry experts estimate that supercomputers will reach 1 EFLOPS (1018, 1,000 PFLOPS or one quintillion FLOPS) by 2018." Which, well, obviously hasn't happened. So should this section be edited or left as a historical note? Bryce Stansfield (talk) 03:19, 10 March 2018 (UTC)

Wiki Education Foundation-supported course assignment
This article was the subject of a Wiki Education Foundation-supported course assignment, between 7 April 2020 and 7 May 2020. Further details are available on the course page. Student editor(s): FarisRaza123.

Above undated message substituted from Template:Dashboard.wikiedu.org assignment by PrimeBOT (talk) 10:27, 17 January 2022 (UTC)

Summit
As of June 2018, Summit (supercomputer) is the fastest. It needs to be added. 5.34.29.116 (talk) 02:15, 9 June 2018 (UTC)

SpiNNaker
I removed the following text from the article:


 * "Spiking Neutral Network Architecture (SpiNNaker),a supercomputer was manufactured by the U.K, unveiled on November 2018( world's largest supercomputer)"

Citation, please? -- The Anome (talk) 18:04, 13 November 2018 (UTC)

Note that we have an article for SpiNNaker already. However, while it's definitely a supercomputer (and how!), it's a quite different beast from other supercomputers: you are unlikely to see SpiNNaker doing LINPACK benchmarks any time soon. I'm pretty sure you could do LINPACK on it by reprogramming the nodes, but that's not the goal of SpiNNaker's design, and I'm sure it would be a waste to time doing it. Perhaps benchmarking in the reverse direction, by comparing neural network simulation performance on conventional supercomputers, might be the best way to compare them? -- The Anome (talk) 10:47, 15 November 2018 (UTC)

Highly biased and hardly related paragraph
I found that a paragraph in this article is highly biased and hardly related to the title "Supercomputer". So I deleted the paragraph. The revision before the deletion Yangbowen1028 (talk) 03:41, 11 March 2019 (UTC)

No, there are no biases here, only undeniable facts! Everything here on this paragraph is a fact and based on reliable sourced references. Even if you don’t like it, It’s still the undeniable truth that the country of Taiwan is the world’s most technologically advanced computer microchip designer and maker. The advanced microchips made by Taiwan serve as the basis of many supercomputers around the world. Please confirm for yourself and read these reference sources:

1.) Taiwan Semiconductor Manufacturing Corporation is most advanced chip maker in the world 2.)Taiwan Semiconductor Manufacturing Corporation beats Intel to become world’s most advanced chip maker — Preceding unsigned comment added by 2600:1011:B02B:31B1:50D4:DFA1:ED42:A144 (talk) 17:46, 11 March 2019 (UTC)

-- Dear Taiwanese National,

Perhaps your paragraph does contain many "undeniable facts." However, they are not pertinent to the wider context of Supercomputers. You would be better including that paragraph in a page on the accomplishments of the Taiwan Technology Sector or your personal blog. The University of Illinois developed many of the early Supercomputers, but you don't see me introducing an entire paragraph about that fact to the introduction to this page.

Signed, - A University of Illinois PhD Candidate.130.126.255.88 (talk) 17:55, 11 March 2019 (UTC)

--

M-13
Sorry, what? Here is some Soveit computer in 1985, that were more powerfull than Cray-2, which is worldwide confirmed to be most powerfull computer in 1985-1987, but nobody never listned anything about this Soviet computer before, in was never mented anywhere by historians and only article about it - it's a this link? Sound very convincing, isn't it? — Preceding unsigned comment added by ZubaZubaRock (talk • contribs) 08:59, 27 March 2019 (UTC)
 * "The Cray-2 was released in 1985. It had eight central processing units (CPUs), liquid cooling and the electronics coolant liquid fluorinert was pumped through the supercomputer architecture. It performed at 1.9 gigaFLOPS and was the world's second fastest after M-13 supercomputer in Moscow.[22="http://www.icfcst.kiev.ua/MUSEUM/Kartsev.html"]"

I groan inside when "high-performance computing" takes me here
Probably about a year ago, I wanted to quickly check out Wikipedia's scope for the industry term "high-performance computing" and my effort took me here instead. I went "ugh" and moved along.

Today I had completely forgotten about this, and again wanted to quickly check out Wikipedia's scope for the industry term "high-performance computing" and again my effort took me here instead. This time, with a shock of sudden recognition, I went "triple ugh" and decided to post a note on the talk page.

Supercomputing has long been a subject area imbued with nationalistic penis wagging. I'm not interested in wading through the annals of jingoistic penis wagging when I want to reference the simple concept of high-performance computing, which to my mind is a set of techniques employed to obtain performance at nearly any effort or expenditure.

Others may have different visceral reactions to this conflation, but this reaction is mine. &mdash; MaxEnt 16:05, 15 September 2020 (UTC)


 * There are over 1400 articles on WP containing the phrase "high-performance computing". I don't know if that helps as far as WP's "scope" for this term.  If you think there should be a separate article, you might want to volunteer to create an article (or perhaps just a stub) or you could request creation of such an article.  In either case, see WP:Article requests. Fabrickator (talk) 19:54, 15 September 2020 (UTC)


 * I restored High-performance computing, that were hidden for 10 years. Today HPC cloud exists (AWS), and not necesarily is a "supercomputer".--GM83 (talk) 21:04, 2 March 2022 (UTC)