鬼子凶猛Japan takes back first place on Top 500 supe ...

来源:百度文库 编辑:超级军网 时间:2024/04/29 09:01:06


转载自:http://www.computerworld.com/s/article/9217755/Japan_takes_back_first_place_on_Top_500_supercomputer_list?taxonomyId=12

A Japanese computer has taken first place on the Top 500 supercomputer list, ending China's reign at the top after just six months. At 8.16 petaflops (quadrillion floating-point calculations per second), the K computer is more powerful than the next five systems combined.

The K computer's performance was measured using 68,544 SPARC64 VIIIfx CPUs each with eight cores, for a total of 548,352 cores, almost twice as many as any other system on the Top500 list. The computer is still under construction, and when it enters service in November 2012 will have more than 80,000 SPARC64 VIIIfx CPUs according to its manufacturer, Fujitsu.

Japan's ascension to the top means that the Chinese Tianhe-1A supercomputer, which took the number 1 position in November last year, is now in second spot with its 2.57 petaflops. But China continues to grow the number of systems it has on the list, up from 42 to 62 systems. The change at the top also means that Jaguar, built for the U.S. Department of Energy (DOE), is bumped down to third place.

The latest iteration of the biannual list was released Monday at the 2011 International Supercomputing Conference in Hamburg, Germany.

Unlike other recent supercomputers, the K computer doesn't use graphics processors or other accelerators. It uses the most power, but is also one of the most energy-efficient systems on the list, according to Top500.org.

The supercomputer is installed at the RIKEN Advanced Institute for Computational Science (AICS) in Kobe. When complete, it is intended to run at over 10 petaflops.

This is the first time Japan has had the most powerful supercomputer since the country's Earth Simulator was surpassed by the DOE's IBM BlueGene/L and by Nasa's Columbia in November 2004.

For the first time, all of the top 10 systems achieved performance over 1 petaflop, although they are the only systems on the list that reach that level. The U.S. has five systems in the top 10; Japan and China have two each, and France has one.

The DOE's Roadrunner, the first system to break the petaflop barrier in June 2008, is now in tenth place. The performance of computers on the list is measured using the Linpack benchmark, a set of routines that solve linear equations.

The last system on the new list was at position 262 six months ago, meaning almost 48 percent of the list has changed in the last six months -- and the turnover rate has steadily increased during the last few lists, according to Top500.org, which publishes the list. While performance at the top is advancing by leaps and bounds, movements lower down the list are more modest. The entry point for the top 100 increased to 88.92 teraflops from 75.76 teraflops six months ago.

IBM is the dominant manufacturer on the list with 213 systems in the Top 500, compared to Hewlett-Packard with 153.

Intel continues to provide the processors for a majority of the systems on list, followed by AMD and IBM. Intel's Westmere processors are now used by 178 systems, up from 56 systems 6 months ago.

The Top 500 list is compiled by Hans Meuer of the University of Mannheim in Germany, Erich Strohmaier and Horst Simon of NERSC/Lawrence Berkeley National Laboratory, and Jack Dongarra of the University of Tennessee, Knoxville.


转载自:http://www.computerworld.com/s/article/9217755/Japan_takes_back_first_place_on_Top_500_supercomputer_list?taxonomyId=12

A Japanese computer has taken first place on the Top 500 supercomputer list, ending China's reign at the top after just six months. At 8.16 petaflops (quadrillion floating-point calculations per second), the K computer is more powerful than the next five systems combined.

The K computer's performance was measured using 68,544 SPARC64 VIIIfx CPUs each with eight cores, for a total of 548,352 cores, almost twice as many as any other system on the Top500 list. The computer is still under construction, and when it enters service in November 2012 will have more than 80,000 SPARC64 VIIIfx CPUs according to its manufacturer, Fujitsu.

Japan's ascension to the top means that the Chinese Tianhe-1A supercomputer, which took the number 1 position in November last year, is now in second spot with its 2.57 petaflops. But China continues to grow the number of systems it has on the list, up from 42 to 62 systems. The change at the top also means that Jaguar, built for the U.S. Department of Energy (DOE), is bumped down to third place.

The latest iteration of the biannual list was released Monday at the 2011 International Supercomputing Conference in Hamburg, Germany.

Unlike other recent supercomputers, the K computer doesn't use graphics processors or other accelerators. It uses the most power, but is also one of the most energy-efficient systems on the list, according to Top500.org.

The supercomputer is installed at the RIKEN Advanced Institute for Computational Science (AICS) in Kobe. When complete, it is intended to run at over 10 petaflops.

This is the first time Japan has had the most powerful supercomputer since the country's Earth Simulator was surpassed by the DOE's IBM BlueGene/L and by Nasa's Columbia in November 2004.

For the first time, all of the top 10 systems achieved performance over 1 petaflop, although they are the only systems on the list that reach that level. The U.S. has five systems in the top 10; Japan and China have two each, and France has one.

The DOE's Roadrunner, the first system to break the petaflop barrier in June 2008, is now in tenth place. The performance of computers on the list is measured using the Linpack benchmark, a set of routines that solve linear equations.

The last system on the new list was at position 262 six months ago, meaning almost 48 percent of the list has changed in the last six months -- and the turnover rate has steadily increased during the last few lists, according to Top500.org, which publishes the list. While performance at the top is advancing by leaps and bounds, movements lower down the list are more modest. The entry point for the top 100 increased to 88.92 teraflops from 75.76 teraflops six months ago.

IBM is the dominant manufacturer on the list with 213 systems in the Top 500, compared to Hewlett-Packard with 153.

Intel continues to provide the processors for a majority of the systems on list, followed by AMD and IBM. Intel's Westmere processors are now used by 178 systems, up from 56 systems 6 months ago.

The Top 500 list is compiled by Hans Meuer of the University of Mannheim in Germany, Erich Strohmaier and Horst Simon of NERSC/Lawrence Berkeley National Laboratory, and Jack Dongarra of the University of Tennessee, Knoxville.
没想到鬼子的速度这么快,居然赶在了Blue Waters之前以一台同构系统达到了8 P的速度...

靠~~难道之前京的“难产”新闻全是忽悠他国的???


SAN FRANCISCO — In the rankings of the world’s most powerful supercomputers, a Japanese machine has earned the top spot with a performance that essentially laps the competition.

The computer, known as “K Computer,” is three times faster than a Chinese rival that previously held the top position, said Jack Dongarra, a professor of electrical engineering and computer science at the University of Tennessee at Knoxville who keeps the official rankings of computer performance.

http://www.nytimes.com/2011/06/20/technology/20computer.html?ref=technology

大体说一下主要内容,日本研制的K型超级计算机最近被评为世界最快的超级计算机。K型超级计算机是由日本富士通公司生产的,计算速度为8.2千万亿次浮点运算/秒,是天河-1A型超级计算机的四倍。
8.2是峰值么天河一号峰值是4700万把好像
谁说超级计算机没用,大家这不都造得轰轰烈烈嘛
奥巴马的马甲 发表于 2011-6-20 18:48
8.2是峰值么天河一号峰值是4700万把好像
原文没说是峰值,

The winning computer was able to make 8.2 quadrillion calculations per second, or in more technical terms, 8.2 petaflops per second.


关于是天河的四倍这一点,原文是这么说的

The computer, known as “K Computer,” is three times faster than a Chinese rival that previously held the top position,
8P Flops……SPARC 64,让龙芯情何以堪啊……
比起阿三的未来型计算机差了好几个数量级


CPU是什么型号??

CPU是什么型号??
最新的超算500榜单出来了,貌似网站挂了,搜到一个榜单,据说是最新的.....

TPimage 发表于 2011-6-20 19:03
最新的超算500榜单出来了,貌似网站挂了,搜到一个榜单,据说是最新的.....
这个就是最新的,纽约时报的报道也是刚刚出来。
失落的天堂 发表于 2011-6-20 18:29
没想到鬼子的速度这么快,居然赶在了Blue Waters之前以一台同构系统达到了8 P的速度...

靠~~难道之前京 ...
大家一起都在造吧,只是鬼子把半成品拿出来露个脸了,目标实际都是20P,IBM未必落后

雅咩嗲·哦耶 发表于 2011-6-20 18:59
CPU是什么型号??


SPARC64 VIIIfx 2000 MHz (16 GFlops)

这是富士通公司特意为K型超级计算机开发的处理器

http://www.itechnews.net/2009/06/09/fujitsu-sparc64-viiifx-venus-worlds-fastest-cpu/

中国的龙芯任重而道远。
雅咩嗲·哦耶 发表于 2011-6-20 18:59
CPU是什么型号??


SPARC64 VIIIfx 2000 MHz (16 GFlops)

这是富士通公司特意为K型超级计算机开发的处理器

http://www.itechnews.net/2009/06/09/fujitsu-sparc64-viiifx-venus-worlds-fastest-cpu/

中国的龙芯任重而道远。
http://tech.qq.com/a/20110620/000332.htm
全球500强超级计算机今出炉 天河1A退居次席
2011年06月20日17:46腾讯科技林靖东我要评论(0)
字号:T|T

全球500强超级计算机今出炉 天河1A退居次席

K Computer(腾讯科技配图)

腾讯科技讯(林靖东)北京时间6月20日消息,据国外媒体报道,国际超级计算机组织今日将在德国汉堡召开2011年年会并公布今年6月份的全球超级计算机500强名单。据称,今年排在超级计算机500强名单首位的超级计算机是日本RIKEN高级计算机科学研究所的K Computer。

这是自2004年以来,日本制造的超级计算机第一次成为全球速度最快的超级计算机。据悉,K Computer的计算性能高达8petaflop/s(千万亿次/秒),远超去年11月时位于榜首的超级计算机。去年11月时排在计算性能排行榜第一位的是中国天津国家超级计算中心的天河-1A,它的计算性能为2.6petaflop/s。

据悉,K Computer是由日本富士通研发而成,包含了8万多个八核心CPU。日本上一次是凭借NEC的Earth Simulator摘得全球最快超级计算机桂冠。在蝉联全球超级计算机500强榜首两年时间后,那款超级计算机于2004年11月被其他超级计算机超越。

在2011年超级计算机500强名单中,排在第二位的是天河-1A,排在第三位的是美国橡树岭国家实验室(Oak Ridge National Laboratory)的美洲豹(Jaquar),它的计算性能为1.75petaflop/s。中国深圳国家超级计算中心的星云(Nebulae)和东京技术研究所(Tokyo Institute of Technology)的燕子2.0(Tsubame 2.0)排在第四位和第五位,它们的计算性能分别为1.27petaflop/s和1.19petaflop/s。

国际超级计算机组织评选超级计算机时使用的标准被称为“线性系统软件包”(LinPack),它是以系统解决线性方程式密集系统的方式来测试系统性能,以每秒进行的计算次数或浮点运算次数作为检测的标准。虽然并非所有人都认为这是对比超级计算机性能的最佳办法,但它是目前应用最广泛的对比方法。

全球超级计算机500强名单更新速度很快。比如,2008年第一个突破petaflop/s级性能的超级计算机系统即洛斯阿拉莫斯国家实验室(Los Alamos National Laboratory)的走鹃(Roadrunner)现在已经掉到了第10位。在2011年全球超级计算机500强名单中,超过petaflop/s级性能的超级计算机系统共有10个。

在全球超级计算机500强名单中,有75台超级计算机被用于科学研究,占比为15%;被应用于金融、服务、万维网和国防的超级计算机数量依次为36台、33台、23台和20台。

从厂商角度来说,IBM仍然是制造全球500强超级计算机数量最多的厂商,大约有42%的超级计算机是由它研发和制造的,然后是惠普和Cray,它们的份额分别为31%和6%。

从国家拥有超级计算机数量的角度来说,美国仍然是拥有500强超级计算机最多的国家,它一共拥有256台500强超级计算机;中国排在第2位,拥有的500强超级计算机数量为62台;随后是德国、英国、日本和法国,这几个国家拥有的超级计算机数量分别为30台、27台、26台和25台。
这样很好,不竞争,哪来的进步。
http://www.riken.jp/engn/r-world/info/release/press/2011/110620/index.html
Supercomputer "K computer" Takes First Place in World
Achieves world's best performance of 8.162 petaflops to lead TOP500 list
PDF: 59KB

June 20, 2011
RIKEN
Fujitsu Limited

Tokyo, June 20, 2011 - RIKEN and Fujitsu have taken first place on the 37th TOP500 list announced today at the 26th International Supercomputing Conference (ISC'11) held in Hamburg, Germany. This ranking is based on a performance measurement of the "K computer,"(*1) currently under their joint development.

The TOP500-ranked K computer system, currently in the configuration stage, has 672 computer racks equipped with a current total of 68,544 CPUs. This half-build system achieved the world's best LINPACK benchmark performance of 8.162 petaflops (quadrillion floating-point operations per second), to place it at the head of the TOP500 list. In addition, the system has recorded high standards with a computing efficiency ratio of 93.0%. This is the first time since June 2004 that the Japanese supercomputer "Earth Simulator" has been ranked first on the TOP500 list.
1. Background

RIKEN and Fujitsu have been working together to develop the K computer, with the aim of beginning shared use by November 2012, as a part of the High-Performance Computing Infrastructure (HPCI) initiative led by Japan's Ministry of Education, Culture, Sports, Science and Technology (MEXT). The K computer will be comprised of over 800 computer racks-each equipped with ultrafast and energy-efficient CPUs-that access into a network capable of an immense amount of interconnectivity. The supercomputer system brings together leading-edge technologies for high performance and high reliability.

To test the system's performance at the configuration stage, the K computer's processing speed was measured by the LINPACK benchmark program, placing it on the 37th TOP500 ranking of the world's fastest supercomputers. The TOP500 ranking list began in 1993 and is updated twice a year in June and November.
2. Performance and Future Status of the K computer

The LINPACK benchmark program, running on the part of the system that employs 68,544 CPUs installed on the K computer being configured, recorded the world's top performance of 8.162 petaflops. This gave it the number-one position on the TOP500 list. Moreover, for one of the world's largest supercomputers, it achieved an extraordinarily high computing efficiency ratio of 93.0%. This achievement is made possible by the K computer's integration of technologies, including its massive number of CPUs, the interconnectivity that links them together, and the software that is able to bring out the highest performance from the hardware.

When configuration of the K computer is complete in 2012, it is designed to achieve LINPACK performance of 10 petaflops. It will be widely used in a variety of computational science fields where it is expected to contribute to the generation of world-class research results. The K computer is a wholly made-in-Japan supercomputer, from the research and development of the processors, to system design and manufacturing. Use of the K computer is expected to have a groundbreaking impact in fields ranging from global climate research, meteorology, disaster prevention, and medicine, thereby contributing to the creation of a prosperous and secure society. RIKEN and Fujitsu will continue to work tirelessly toward completing the system's deployment in 2012.
3. RIKEN and Fujitsu Comments

Ryoji Noyori, President, RIKEN
I would like to express my deep gratitude to everyone, beginning with our colleagues at our development partner Fujitsu Limited, who worked so valiantly on the construction of the K computer even under the severe conditions following the Great East Japan Earthquake. It is wonderful to be able to share the joy of this moment with them. I very much believe that the strength and perseverance that was demonstrated during this project will also make possible the recovery of the devastated Tohoku region. As we move forward to complete this project by next June, we will maintain our firm commitment to the maintenance and operation of the system, and I hope to see wonderful results when we begin to make the world's top performing supercomputer available to users around the world.

Michiyoshi Mazuka, Chairman and Representative Director, Fujitsu Limited
I am delighted that we were able to achieve this result, made possible through the tremendous efforts of all involved, despite the impact of the Great East Japan Earthquake. In particular, I am sincerely grateful to our partners in the Tohoku region for their commitment to delivering a steady supply of components, even though they themselves were affected by the disaster. Bringing together hundreds of thousands of components to quickly launch such a massive-scale computing system-which would have been nearly impossible using conventional technologies-requires an incredible level of reliability. I believe that this reliability is truly the pinnacle of Japanese manufacturing. Without being too pleased with ourselves and losing sight of our goal, going forward we will proceed with the system's deployment and, once complete, we look forward to contributing to the achievements that the K computer will make possible.

Press Contacts
    RIKEN
    Advanced Institute for Computational Science
    Office for Research Communications
    Tel: +81-78-940-5623, 5624
    Fujitsu Limited
    Public and Investor Relations Division
    InquiriesNew Window

Page Top

Supplementary Explanation
1        K computer
The K computer, which is being jointly developed by RIKEN and Fujitsu, is part of the High-Performance Computing Infrastructure (HPCI) initiative led by Japan's Ministry of Education, Culture, Sports, Science and Technology (MEXT). Configuration of the K computer began in the end of September 2010, with availability for shared use scheduled for 2012. The "K computer" is the nickname RIKEN has been using for the supercomputer of this project since July 2010. "K" comes from the Japanese Kanji letter "Kei" which means ten peta or 10 to the 16th power. The logo for the K computer based on the Japanese letter for Kei, was selected in October 2010. In its original sense, "Kei" expresses a large gateway, and it is hoped that the system will be a new gateway to computational science.
2        LINPACK
A program developed by J. Dongarra, Ph.D., of the University of Tennessee, for solving a system of linear equations using matrix computation. It is the benchmark program used to create the TOP500 list, which ranks the performance of the world's supercomputers (announced in June and November of every year).
Page Top

About RIKEN

The mission of RIKEN is to conduct comprehensive research in science and technology (excluding only the humanities and social sciences) as provided for under the "RIKEN Law," and to publicly disseminate the results of its scientific research and technological developments. RIKEN carries out high level experimental and research work in a wide range of fields, including physics, chemistry, medical science, biology, and engineering, covering the entire range from basic research to practical application. RIKEN was first organized in 1917 as a private research foundation, and reorganized in 2003 as an independent administrative institution under the Ministry of Education, Culture, Sports, Science and Technology.

About Fujitsu

Fujitsu is a leading provider of information and communication technology (ICT)-based business solutions for the global marketplace. With approximately 170,000 employees supporting customers in over 100 countries, Fujitsu combines a worldwide corps of systems and services experts with highly reliable computing and communications products and advanced microelectronics to deliver added value to customers. Headquartered in Tokyo, Fujitsu Limited (TSE:6702) reported consolidated revenues of 4.5 trillion yen (US$55 billion) for the fiscal year ended March 31, 2011. For more information, please see: www.fujitsu.comNew Window.
Figure 1

Figure 2

The K computer as it looks today
Figure 3

A K computer rack on display at the RIKEN Advanced Institute for Computational Science
Figure 4

The press conference on June 20. From left: RIKEN Advanced Institute for Computational Science Director Kimihiko Hirao, RIKEN President Ryoji Noyori, Fujitsu Limited Chairman and Representative Director Michiyoshi Mazuka, Fujitsu Limited Next Generation Technical Computing Unit President Aiichiro Inoue

机翻:

超级计算机“K电脑”注意到在世界第一位
达到世界上最好的8.162 petaflops的性能领导TOP500名单
PDF:59KB

2011年6月20日
RIKEN
富士通公司

东京6月20日,2011 -日本理化学研究所和富士通已在第37届TOP500排行榜首位宣布今天在第26届国际超级计算机大会在德国汉堡举行(ISC'11)。 这个排名是根据他们目前正在联合开发的“K电脑”(* 1)的性能测量。

在TOP500排名K电脑系统,目前在配置阶段,已与672电脑的处理器目前共有68544架装备。 这半生成系统实现了世界上最好的8.162 petaflops的LINPACK基准性能(万亿浮点每秒运算),放置在500强名单头上。 此外,该系统已录得的93.0%,与计算效率比的高标准。 这是自2004年6月首次在日本超级计算机“地球模拟器”已名列TOP500排行榜首位。
1。 背景

日本理化学研究所和富士通已合作开发了K电脑,在开始共享2012年11月使用的目的,作为高性能计算基础设施(HPCI)由日本教育部,文化,体育,科学主导计划的一部分与技术(文部科学省)。 在K电脑将包括800多个电脑架,每超高速和高能效处理器,即成为一个网络的互联互通能够访问量巨大的配备。 该超级计算机系统汇集了高性能和高可靠性的先进技术。

为了测试系统的配置在舞台表演中,K计算机的处理速度是衡量的LINPACK基准程序,放置在第37届TOP500是世界上最快的超级计算机排名的。 在500强排行榜始于1993年,是更新六月和十一月两次。
2。 性能和K电脑未来地位

在LINPACK基准测试程序,对系统所配置上,它采用了K 68544的CPU的计算机上安装的一部分运行,记录了世界的8.162 petaflops的最佳性能。 这给了它的头号TOP500名单上的位置。 此外,为世界上最大的超级计算机之一,它实现了93.0%的比例计算效率非常高。 这一成就成为可能的K电脑的技术的集成,包括其数量庞大的CPU,在互联互通的联系在一起,而软件即能衬托出从硬件性能最高的。

当计算机配置的K是在2012年完成,它是旨在实现10 petaflops的LINPACK性能。 这将是广泛用于计算科学的地方,预计有助于世界一流的研究成果产生各种领域。 在K计算机是完全制成的日本超级计算机,从研究和开发的处理器,系统设计和制造。 使用计算机的K,预计有来自全球气候研究,气象,灾害预防和医学领域的开创性不等的影响,从而促进了繁荣和安全的社会。 日本理化学研究所和富士通将继续朝着2012年完成系统的部署而不懈努力。
3。 日本理化学研究所和富士通评论

野依良治,主席,日本理化学研究所
我想表达我深深的感谢大家,我们的同事们开始在我们的发展伙伴富士通公司,谁曾在计算机的K建设,勇敢,即使在下面大东日本地震的严重状况。 它是美好的,能够与他们分享这一时刻的喜悦。 我很相信的力量和毅力的是在这个项目中表现也将成为可能的破坏东北地区的复苏。 当我们向前迈进,完成明年六月这个项目,我们将保持我们对维护和系统运行的坚定承诺,我希望能看到精彩的结果,当我们开始使世界顶级性能的超级计算机提供给世界各地的用户。

Michiyoshi玛祖卡,主席和代表董事,富士通有限公司
我很高兴我们能够达到这个结果,通过了涉及所有可能的巨大努力,尽管大东日本地震的影响。 特别是,我衷心感谢在他们承诺提供的组件的稳定供应东北地区的合作伙伴,虽然他们自己也受灾。 汇集了成千上万的组件迅速启动如此庞大规模的计算系统,这将是几乎不可能使用传统的技术,需要的可靠性令人难以置信的程度。 我认为,这是真正的可靠性,日本制造业的巅峰。 没有过于高兴,我们自己和我们的目标失去视力,展望未来,我们将继续进行系统的部署,一旦完成,我们期待着贡献的成就,在K计算机将成为可能。

媒体联系方式
    RIKEN
    先进的计算科学研究所
    办公室研究通讯
    电话:+81-78-940-5623,5624
    富士通公司
    公众和投资者关系部
    咨询 新窗口

页首

补充说明
1         K电脑
在K电脑,目前正由日本理化学研究所和富士通开发,是高性能计算基础设施(HPCI)由日本教育部,文化,体育,科学和技术(文部科学省)领导倡议的一部分。 电脑配置的K开始在2010年9月底,与2012年定为共同使用的可用性。 在“K电脑”是绰号理化学研究所已为这个项目的超级计算机采用自2010年7月。 “K”来自日本的汉字字母“基”,这意味着ten PETA或10到16的力量。 为在日本的基信为本的K计算机标识,被选为2010年10月。 在其原有的意义上说,“箕”表达了一种大型的门户,并希望该系统将是一个新的门户,计算科学。
2         LINPACK
由J. Dongarra博士开发出一种方案,美国田纳西州大学,解决了利用矩阵计算线性方程组。 这是基准程序用于创建500强名单,位列世界超级计算机(6月和11月宣布每年)的性能。
页首

关于理化学研究所

理化学研究所的使命是进行科学和技术的综合研究(不包括只有人文和社会科学)作为下提供的“理化学研究所法”,并公开传播其科学研究和技术发展的结果。 日本理化学研究所进行的实验中高层次的领域广泛,包括物理,化学,医学,生物,工程,覆盖了整个范围从基础研究到实际应用和研究工作。 日本理化学研究所首次举办于1917年作为一个私人研究的基础,作为一个独立的行政机构改组于2003年根据教育部,文化,体育,科学和技术。

关于富士通

富士通是信息和通信技术(ICT)为基础的商业解决方案的全球市场领先供应商。 拥有约170000名员工在100多个国家支持的客户,富士通结合的系统和高度可靠的计算通讯产品和先进的微电子技术,提供增值服务的客户遍布世界各地的专家。 总部设在东京,富士通有限公司(TSE:6702)公布了截至2011年3月31日的财年4.5万亿日元(550亿美元)的综合收入。 欲了解更多信息,请参阅: www.fujitsu.com 新窗口 。
图1

图2

在K电脑,因为它看起来今天
图3

AK计算机机架上的理化学研究所高级研究所显示计算科学
图4

6月20日的新闻发布会。 左起:理化学研究所高级研究所计算科学主任Kimihiko平尾,野依良治理化学研究所总裁,富士通有限公司的主席及代表董事Michiyoshi玛祖卡,富士通公司的下一代技术计算部总裁Aiichiro井上
貌似倭国最牛X的地方在于实现了93%的计算效率,天河一号1A还不到60%的计算效率
http://top500.org/lists/2011/06/press-release

Japan Reclaims Top Ranking on Latest TOP500 List of World’s Supercomputers
Thu, 2011-06-16 19:24

Contact: Erich Strohmaier, EStrohmaier@lbl.gov, 510-495-2517

HAMBURG, Germany—A Japanese supercomputer capable of performing more than 8 quadrillion calculations per second (petaflop/s) is the new number one system in the world, putting Japan back in the top spot for the first time since the Earth Simulator was dethroned in November 2004, according to the latest edition of the TOP500 List of the world’s top supercomputers. The system, called the K Computer, is at the RIKEN Advanced Institute for Computational Science (AICS) in Kobe.

The 37th edition of the closely watched list was released Monday, June 20, at the 2011 International Supercomputing Conference in Hamburg. The ranking of all systems is based on how fast they run Linpack, a benchmark application developed to solve a dense system of linear equations.

For the first time, all of the top 10 systems achieved petaflop/s performance – and those are also the only petaflop/s systems on the list. The U.S. is tops in petaflop/s with five systems performing at that level; Japan and China have two each, and France has one.

Bumped to second place after capturing No. 1 on the previous list is the Tianhe-1A supercomputer the National Supercomputing Center in Tianjin, China, with a performance at 2.6 petaflop/s. Also moving down a notch was Jaguar, a Cray supercomputer at the U.S. Department of Energy’s (DOE’s) Oak Ridge National Laboratory, at No. 3 with 1.75 petaflop/s.

Rounding out the Top 10 are Nebulae at China’s National Supercomputing Center in Shenzen (1.27 petaflop/s), Tsubame 2.0 at the Tokyo Institute of Technology (1.19 petaflop/s), Cielo at Los Alamos National Laboratory in New Mexico (1.11 petaflop/s), Pleiades at the NASA Ames Research Center in California (1.09 petaflop/s), Hopper at DOE’s National Energy Research Scientific Computing Center (NERSC) in California (1.054 petaflop/s), Tera 100 at the CEA (Commissariat à l'énergie atomique et aux énergies alternatives) in France (1.05 petaflop/s), and Roadrunner at Los Alamos National Laboratory in New Mexico (1.04 petaflop/s).

The New Number One

The K Computer, built by Fujitsu, currently combines 68544 SPARC64 VIIIfx CPUs, each with eight cores, for a total of 548,352 cores—almost twice as many as any other system in the TOP500. The K Computer is also more powerful than the next five systems on the list combined.

The K Computer’s name draws upon the Japanese word "Kei" for 10^16 (ten quadrillions), representing the system's performance goal of 10 petaflops. RIKEN is the Institute for Physical and Chemical Research. Unlike the Chinese system it displaced from the No. 1 slot and other recent very large system, the K Computer does not use graphics processors or other accelerators. The K Computer is also one of the most energy-efficient systems on the list.
Some Other Highlights from the newest List

    The two Chinese systems at No. 2 and No. 4 and the Japanese Tsubame 2.0 system at No. 5 are all using NVIDIA GPUs to accelerate computation, and a total of 19 systems on the list are using GPU technology.
    China keeps increasing its number of systems and is now up to 62, making it clearly the No. 2 country as a user of HPC, ahead of Germany, UK, Japan and France.
    Intel continues to provide the processors for the largest share (77.4 percent) of TOP500 systems. Intel’s Westmere processors increased their presence in the list strongly with 169 systems, compared with 56 in the last list.
    Quad-core processors are used in 46.2 percent of the systems, while already 42.4 percent of the systems use processors with six or more cores.
    Cray defended the No. 2 spot in market share by total against Fujitsu, but IBM stays well ahead of either. Cray’s XT system series remains very popular for big research customers, with three systems in the TOP 10 (one new and two previously listed).

Tracking Changes

Since its launch in 1993, the TOP500 List has provided a consistent metric for evaluating the performance of supercomputers. This consistency has made the list a valuable tool for tracking changes in the industry, both over the years and from list to list. Here are some notable changes between the current list and the 36th edition, issued in November 2010.

    The entry level to the list moved up to the 40.1 Tflop/s mark on the Linpack benchmark, compared to 31.1 Tflop/s six months ago.
    The last system on the newest list was listed at position 262 six months ago. This turnover rate has steadily increased during the last few lists and is now above average.
    Total combined performance of all 500 systems has grown to 58.88 Pflop/s, compared to 43.7 Pflop/s six months ago and 32.4 Pflop/s one year ago.
    The average concurrency level in the TOP500 is 15,550 cores per system, up from 13,071 six months ago and 10,267 one year ago.
    The U.S. is the leading consumer of HPC systems with 256 of the 500 systems (down from 274). The European share (125 systems ) is still larger than the Asian share (103 systems – up from 84). Dominant countries in Asia are China with 62 systems (up from 42) and Japan with 26 systems (unchanged). In Europe, Germany, UK, and France are almost equal with 30 (26), 27 (24), and 25 (26) respectively.

Some Final Notes on Power Consumption

Just as the TOP500 List has emerged as a standardized indicator of performance and architecture trends since it was created 18 years ago, the list now tracks actual power consumption of supercomputers in a consistent fashion. Although power consumption is increasing, the computing efficiency of the systems is also improving. Here are some power consumption notes from the newest list.

    29 systems on the list are confirmed to use more than 1 megawatt (MW) of power.
    Average power efficiency is 248 Mflops/watt (up from 219 Mflops/watt six months ago and 195 Mflops/watt one year ago).
    The No. 1 system, the K Computer, also reports the highest total power consumption of 9.89 MW.
    Average power consumption of a TOP500 system is 543 KW (up from 447 KW six months ago and 397 KW one year ago).
    Average power consumption of a TOP 10 system is 4.3 MW (up from 3.2 MW six months ago), and average power efficiency is 464 Mflops/watt (up from 268 Mflops/watt six months ago).

About the TOP500 List

The TOP500 list is compiled by Hans Meuer of the University of Mannheim, Germany; Erich Strohmaier and Horst Simon of NERSC/Lawrence Berkeley National Laboratory; and Jack Dongarra of the University of Tennessee, Knoxville

机翻:
日本回收的最新世界500强的超级计算机名单热门排行
周四,2011年6月16日19:24

联系人:的Erich Strohmaier, EStrohmaier@lbl.gov ,510-495-2517

汉堡,德国一位日本超级计算机的性能超过8万亿计算每秒(petaflop的/ s)的能力是新的头号系统在世界上,在把交椅日本回来的第一次,因为在地球模拟器被废黜2004年11月,根据世界顶级超级计算机500强名单的最新版本。 该系统称为计算机的K,是在日本理化学研究所高级研究所计算科学(AICS)在神户​​。

而备受关注的名单第37版于周一公布,6月20日,在2011年汉堡国际超级计算机会议。 所有系统的排名是根据他们运行的Linpack速度有多快,基准应用程序开发,解决了线性方程密集系统。

这是第一次,前10个系统全部实现千万亿次/秒的性能 - 而这些都是名单上的也只有千万亿次/ s系统。 美国是在千万亿次/含五国在该级别执行系统s上衣,日本和中国有两个,每年,法国有一个。

被撞后,捕捉到第二位在前面的列表1号是天河- 1A超级计算机在天津,中国国家超级计算机中心的性能,在2.6千万亿次/秒。 同时向下移动一个档次了捷豹,在美国能源部的美国能源部(DOE的)橡树岭国家实验室的Cray超级计算机排在第3,用1.75千万亿次/秒。

列前十位的是星云在中国的深圳国家超级计算中心(1.27千万亿次/秒),燕2.0在东京技术研究所(1.19千万亿次/秒),在洛斯阿拉莫斯国家实验室在新墨西哥州CIELO(1.11千万亿次/秒),在美国航天局艾姆斯研究中心的加利福尼亚州(1.09千​​万亿次/秒),霍普在美国能源部的国家能源研究科学计算中心(NERSC)在加利福尼亚州(1.054千万亿次/秒),在100万亿CEA(昴小卖部A L' ENERGIE atomique等辅助能源的替代品),法国(1.05千万亿次/秒),并在洛斯阿拉莫斯国家实验室在新墨西哥州(1.04千万亿次/秒)越野车。

新一号

在K电脑,富士通建成,目前结合为548,352总68544 SPARC64 VIIIfx的CPU,有八个内核,每个内核,内核,几乎两倍TOP500中的任何​​其他系统的许多。 在K电脑也超过了未来五年的组合列表系统的强大。

在K计算机的名称借鉴日本字“箕”10 ^ 16(十quadrillions),占系统的10 petaflops的性能目标。 RIKEN是物理和化学研究所。 不像中国的制度是从1号插槽和其他流离失所最近非常大的系统,在K电脑无法使用图形处理器或其他加速器。 在K电脑也是名单上的最节能的系统之一。
从最新的名单一些其他亮点

    这两个在第2和第4号和日本燕2.0在第5系统中国系统都使用NVIDIA GPU来加速计算,以及19名单上的系统的总使用GPU技术。
    中国不断增加其系统的数量,现在到62,使得作为一个HPC用户清楚的2号国,领先于德国,英国,日本和法国。
    英特尔将继续提供最大份额的500强系统(77.4%)的处理器。 英特尔Westmere的处理器,增加了与169系统列表强烈的存在,而在过去的清单56。
    四核处理器采用的系统在46.2%,而已有42.4%的用户使用的系统有六个或更多内核的处理器。
    克雷辩护,对富士通总的市场份额排名第二,但IBM保持良好的要么前进。 Cray的XT系统系列仍非常大研究客户的青睐,有三个前10名系统(一个新的,上面列出的两个)。

跟踪变更

自1993年推出以来,500强名单已经为评价超级计算机的性能一致的指标。 这种一致性使列表中的一个在行业跟踪变化的重要工具,无论是过去数年,从清单列表。 下面是在当前列表和第36版一些显着变化,2010年11月发行。

    入门级到列表上升到40.1 Tflop /在Linpack基准S标志,而31.1 Tflop / s的半年前。
    最新的名单上列出的最后一个系统是在半年前262位。 这个周转率稳步上升,在过去数列出现在是高于平均水平。
    共有500个系统的所有组合的表现已发展到58.88 Pflop /秒,比43.7 Pflop / s的半年前和32.4 Pflop / S一年前。
    500强中的平均水平是15,550并发每个系统的核心,从13071和10267六个月前一年前。
    美国是领先的高性能计算系统与500系统(低于274)256消费者。 欧洲的份额(125系统)仍然比亚洲的份额(103系统 - 从84)大。 在亚洲的主导国家是中国与62系统(从42)和26个系统(不变)日本。 在欧洲,德国,英国和法国几乎与30(26),27(24),25(26)分别相等。

电力消耗一些最后的注意事项

正如500强名单已作为一个标准化的性能和结构趋势的指标出现,因为它是建立在18年前,现在的名单跟踪以一致的方式超级计算机的实际功耗。 虽然功率消耗增加,对系统计算效率也在提高。 下面是一些从最新名单功耗笔记。

    29名单上的系统被证实使用超过1兆瓦的功率(MW)。
    平均功率效率是248 MFLOPS /瓦(从219 MFLOPS /瓦半年前和195 MFLOPS /瓦一年前)。
    1号系统,在K电脑,还报告了最高的9.89兆瓦的总功率消耗。
    TOP500系统的平均功耗是543千瓦(447千瓦从半年前和一年前397千瓦)。
    TOP 10系统的平均耗电量为4.3兆瓦(从3.2兆瓦半年前),平均功耗效率是464 MFLOPS /瓦(从268 MFLOPS /瓦半年前)。

关于500强名单

在500强名单是由Hans Meuer的德国曼海姆大学的Erich Strohmaier和Horst NERSC /劳伦斯伯克利国家实验室的西蒙和杰克的田纳西大学诺克斯维尔Dongarra。
热烈欢迎超级计算机领域的军备竞赛


top500有个比较表格,下载地址是:http://top500.org/static/lists/2011/06/TOP500_201106.xls

内容部分截图如下;



峰值貌似是8.77petaflops per second.....




top500有个比较表格,下载地址是:http://top500.org/static/lists/2011/06/TOP500_201106.xls

内容部分截图如下;

11111.jpg (127.4 KB, 下载次数: 1)

下载附件 保存到相册

2011-6-20 19:46 上传



峰值貌似是8.77petaflops per second.....



TPimage 发表于 2011-6-20 19:29
貌似倭国最牛X的地方在于实现了93%的计算效率,天河一号1A还不到60%的计算效率


相比起日本来说,我认为我们差距最大还是芯片的自给能力,天河-1A基本上就是由国外公司的芯片组装起来的,而日本的K型可以算是真正的日本产
TPimage 发表于 2011-6-20 19:29
貌似倭国最牛X的地方在于实现了93%的计算效率,天河一号1A还不到60%的计算效率


相比起日本来说,我认为我们差距最大还是芯片的自给能力,天河-1A基本上就是由国外公司的芯片组装起来的,而日本的K型可以算是真正的日本产
http://it.sohu.com/20110620/n310921891.shtml

最新超级计算机500强榜单发布 日本K折桂
来源:搜狐IT
2011年06月20日19:04


  【搜狐IT消息】北京时间6月20日消息,据国外媒体报道,周一公布的最新的全球超级计算机500强排行榜显示,日本的K超级计算机的运算能力为每秒8.16千万亿次,超过中国的天号-1A成为全球运算能力最强的超级计算机系统。

  K配置68544个SPARC64 VIIIfx CPU(中央处理器),每个CPU集成有8个内核,内核总数为548352个,几乎相当于排行榜上其他系统的2倍。据制造商富士通称,K的制造尚未完成,2012年11月份完成制造后,K将配置80000个SPARC64 VIIIfx CPU,运算能力将超过每秒10千万亿次。

  天号-1A以每秒2.57千万亿次的运算能力排在第二位。但是,中国上榜的超级计算机数量由半年前的42台增长到62台;美国能源部的美洲虎(Jaguar)的排名则下滑到第三位。

  与其他最近研发的超级计算机不同,K没有使用图形芯片等加速处理器。K的能耗最高,但它也是榜单上效能比最高的超级计算机系统之一。

  在最新的超级计算机500强榜单上,IBM系为213台,惠普系为153台;按芯片分,排在前三位的依次是英特尔、AMD和IBM。(阳光)
nimo.cn 发表于 2011-6-20 19:16
SPARC64 VIIIfx 2000 MHz (16 GFlops)

这是富士通公司特意为K型超级计算机开发的处理器
中国现在又GPU研究计划吗
就这样 老牌强国 不过理研这个K字 搞不好韩国人。。。
北京时间6月20日消息,据国外媒体报道,周一公布的最新的全球超级计算机500强排行榜显示,日本的K超级计算机的运算能力为每秒8.16千万亿次,超过中国的天号-1A成为全球运算能力最强的超级计算机系统。K配置68544个SPARC64 VIIIfx CPU(中央处理器),每个CPU集成有8个内核,内核总数为548352个,几乎相当于排行榜上其他系统的2倍。

  据制造商富士通称,K的制造尚未完成,2012年11月份完成制造后,K将配置80000个SPARC64 VIIIfx CPU,运算能力将超过每秒10千万亿次。

  天号-1A以每秒2.57千万亿次的运算能力排在第二位。但是,中国上榜的超级计算机数量由半年前的42台增长到62台;美国能源部的美洲虎(Jaguar)的排名则下滑到第三位。

  与其他最近研发的超级计算机不同,K没有使用图形芯片等加速处理器。K的能耗最高,但它也是榜单上效能比最高的超级计算机系统之一。

  在最新的超级计算机500强榜单上,IBM系为213台,惠普系为153台;按芯片分,排在前三位的依次是英特尔、AMD和IBM。
http://roll.sohu.com/20110620/n310927268.shtml
奥巴马的马甲 发表于 2011-6-20 18:48
8.2是峰值么天河一号峰值是4700万把好像
它的峰值和实际差不多 这点很靠谱
曙光6000是怎么个情况?
今年中国应该会发布2台,等着看吧。
nimo.cn 发表于 2011-6-20 18:45
SAN FRANCISCO — In the rankings of the world’s most powerful supercomputers, a Japanese machin ...
“three times faster”咋被翻译成快四倍了呢?

deam 发表于 2011-6-20 18:57
8P Flops……SPARC 64,让龙芯情何以堪啊……


张立新博士现任中科院计算所研究员、博士生导师、副总工程师,中组部“千人计划”入选者。 1993 年毕业于复旦大学,2001 年获美国犹他大学计算机科学博士学位。2001 年至 2003 年在犹他大学做博士后,与美国SGI 公司合作进行千万亿次超级计算机前瞻性研究工作。2003.9-2010.7在美国国际商用机器公司研究院(IBM Research)工作,主要从 事计算机体系结构和微结构、高性能计算机、性能分析、和体系结构模拟器等领域的研发。 主要科研经历和取得的研究成果有:

1、作为IBM 万万亿次机研发项目PERCS 的主要成员之一,参与了Cell 和Power7 芯片 的设计。
2、作为IBM 千万亿次高效计算机项目Percs 和Blue Waters 万万亿次超级计算机项目 的主管设计者之一,首席模拟器/性能架构师,该系统将于2011 年中期问世。
3、作为IBM “无所不在的高性能计算(UHPC)”项目首席芯片架构师,主要研究2017 年左右百万万亿次超级计算机、千万亿次机柜、万亿次嵌入式系统的关键技术,负责处 理器芯片的设计。
4、作为IBM 3D 芯片研究项目中两个领头系统架构师之一,主要研究可用于未来3D 芯 片的高效系统结构。已发表了30 多篇学术论文,所发表论文已被引用了600 多次,申请 了60多项发明专利,于2009 年评为IBM 大师级发明家。20 多次担任国际会议程序委员 会委员等职务,多次获IBM 奖励,其中包括2 次个人杰出技术成就奖,3 次技术成就奖,3 次喝彩奖, 15 次发明奖和18 次感谢奖。



deam 发表于 2011-6-20 18:57
8P Flops……SPARC 64,让龙芯情何以堪啊……


张立新博士现任中科院计算所研究员、博士生导师、副总工程师,中组部“千人计划”入选者。 1993 年毕业于复旦大学,2001 年获美国犹他大学计算机科学博士学位。2001 年至 2003 年在犹他大学做博士后,与美国SGI 公司合作进行千万亿次超级计算机前瞻性研究工作。2003.9-2010.7在美国国际商用机器公司研究院(IBM Research)工作,主要从 事计算机体系结构和微结构、高性能计算机、性能分析、和体系结构模拟器等领域的研发。 主要科研经历和取得的研究成果有:

1、作为IBM 万万亿次机研发项目PERCS 的主要成员之一,参与了Cell 和Power7 芯片 的设计。
2、作为IBM 千万亿次高效计算机项目Percs 和Blue Waters 万万亿次超级计算机项目 的主管设计者之一,首席模拟器/性能架构师,该系统将于2011 年中期问世。
3、作为IBM “无所不在的高性能计算(UHPC)”项目首席芯片架构师,主要研究2017 年左右百万万亿次超级计算机、千万亿次机柜、万亿次嵌入式系统的关键技术,负责处 理器芯片的设计。
4、作为IBM 3D 芯片研究项目中两个领头系统架构师之一,主要研究可用于未来3D 芯 片的高效系统结构。已发表了30 多篇学术论文,所发表论文已被引用了600 多次,申请 了60多项发明专利,于2009 年评为IBM 大师级发明家。20 多次担任国际会议程序委员 会委员等职务,多次获IBM 奖励,其中包括2 次个人杰出技术成就奖,3 次技术成就奖,3 次喝彩奖, 15 次发明奖和18 次感谢奖。


The computer is still under construction, and when it enters service in November 2012
大家不觉得很有喜感吗?
LPC2103 发表于 2011-6-21 10:24
The computer is still under construction, and when it enters service in November 2012
大家不觉得很 ...
是不是已经能提供部分运算服务了?
LPC2103 发表于 2011-6-21 10:24
The computer is still under construction, and when it enters service in November 2012
大家不觉得很 ...
这有啥好喜感的,天河不也是登榜很长时间后才开始提供计算服务的。
deam 发表于 2011-6-21 11:16
这有啥好喜感的,天河不也是登榜很长时间后才开始提供计算服务的。
still under construction,天河1A是吗?
LPC2103 发表于 2011-6-21 11:26
still under construction,天河1A是吗?
不是还要继续扩展计算能力么。等建造完毕,计算速度也不止8P了。
deam 发表于 2011-6-21 11:28
不是还要继续扩展计算能力么。等建造完毕,计算速度也不止8P了。
继续扩展计算能力和still under construction是一个概念吗?

deam 发表于 2011-6-21 11:16
这有啥好喜感的,天河不也是登榜很长时间后才开始提供计算服务的。


天河一号A是天河一号的扩充,2010年11月的时候,天河已经提供服务了,创纪录的是天河一号A。
现在不太清楚,K是不是已经能提供服务了。

天河一号
  –6144个通用处理器(3072x2 Intel Quad Core Xeon E5540 2.53GHz/E5540 3.0GHz);
  –5120个加速处理器(2560 ATI Radeon 4870x2 575MHz)
  –内存总容量98TB ;
  –点点通信带宽40Gbps ;
  –共享磁盘总容量为1PB
天河一号A
  –2048颗我国仿制sun公司的UltraSparc T2处理器(飞腾FT-1000)八核心处理器
  –14336颗Intel Xeon X5670 2.93GHz六核心处理器
  –7168块NVIDIA Tesla M2050高性能计算卡

deam 发表于 2011-6-21 11:16
这有啥好喜感的,天河不也是登榜很长时间后才开始提供计算服务的。


天河一号A是天河一号的扩充,2010年11月的时候,天河已经提供服务了,创纪录的是天河一号A。
现在不太清楚,K是不是已经能提供服务了。

天河一号
  –6144个通用处理器(3072x2 Intel Quad Core Xeon E5540 2.53GHz/E5540 3.0GHz);
  –5120个加速处理器(2560 ATI Radeon 4870x2 575MHz)
  –内存总容量98TB ;
  –点点通信带宽40Gbps ;
  –共享磁盘总容量为1PB
天河一号A
  –2048颗我国仿制sun公司的UltraSparc T2处理器(飞腾FT-1000)八核心处理器
  –14336颗Intel Xeon X5670 2.93GHz六核心处理器
  –7168块NVIDIA Tesla M2050高性能计算卡

LPC2103 发表于 2011-6-21 11:29
继续扩展计算能力和still under construction是一个概念吗?


可top500排行榜又不管你是在建还是已经提供服务,只看你能不能跑下linpack测试而已。天河1号不也是2009年10月发布,1年后才开始服务的。
LPC2103 发表于 2011-6-21 11:29
继续扩展计算能力和still under construction是一个概念吗?


可top500排行榜又不管你是在建还是已经提供服务,只看你能不能跑下linpack测试而已。天河1号不也是2009年10月发布,1年后才开始服务的。
LPC2103 发表于 2011-6-21 10:24
The computer is still under construction, and when it enters service in November 2012
大家不觉得很 ...
搜狐新闻中说了:

K配置68544个SPARC64 VIIIfx CPU(中央处理器),每个CPU集成有8个内核,内核总数为548352个,几乎相当于排行榜上其他系统的2倍。据制造商富士通称,K的制造尚未完成,2012年11月份完成制造后,K将配置80000个SPARC64 VIIIfx CPU,运算能力将超过每秒10千万亿次。

----倭国的K computer就是奔着10千万亿次去的,目前还没完工,跑一遍linpack测试就创了记录了.......
     看看下半年TG能扳回这局不?