POPPUR爱换

标题: 天河一号 A 被日本 “京” 爆掉 [打印本页]

作者: Edison 时间: 2011-6-20 19:42
标题: 天河一号 A 被日本 “京” 爆掉
http://www.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.

作者: snapstop 时间: 2011-6-20 19:43
老大，我承认我很懒，简单概括下吧

作者: gtx5 时间: 2011-6-20 19:52
没有用GPU就能爆天河{:5_196:}

作者: voidshatter 时间: 2011-6-20 19:59
我擦，没有用显卡，超过天河1A三倍的运算力，效率还超过90%

作者: 66666 时间: 2011-6-20 21:00
本帖最后由 66666 于 2011-6-20 21:08 编辑

接近7万个Sparc64 VIIIfx，45nm下跟费米大小差不多，性能大约有128Gflops，日本人还真是变态花这么高的代价。。。。

作者: 左脚选手 时间: 2011-6-20 21:12

voidshatter 发表于 2011-6-20 19:59
我擦，没有用显卡，超过天河1A三倍的运算力，效率还超过90%

CPU的效率就是高，这个没办法，现在SPARC CPU应该都是富士通生产了吧，也算日本自己生产的CPU，这点比我们拿intel的CPU和NV的GPU组装的天河强。

作者: Buffer 时间: 2011-6-20 21:16
本帖最后由 Buffer 于 2011-6-20 21:28 编辑

我理解错了

作者: luckissy 时间: 2011-6-20 21:25
{lol:]全是自产的

作者: disruptor 时间: 2011-6-20 22:16
8p,至少**是第一个使用gpu拿到第一的

作者: potomac 时间: 2011-6-20 22:35
提示: 作者被禁止或删除内容自动屏蔽

作者: asdfjkl 时间: 2011-6-20 22:39
本帖最后由 asdfjkl 于 2011-6-20 22:40 编辑

Edison 发表于 2011-6-20 19:42
http://www.top500.org/lists/2011/06/press-release

我算了算，每个CPU的计算能力也不到150GFlop/s; 龙芯都有这个成绩把，哈哈！
用了68544 SPARC64 VIIIfx CPUs，这真的不计代价的堆数量堆出来的；
堆这个数量，你用Intel已经出来的或者AMD推土机的，成绩只会比这个更好。
小日本也不把报价说说，造价就是它的软肋！

作者: asdfjkl 时间: 2011-6-20 22:46

66666 发表于 2011-6-20 21:00
接近7万个Sparc64 VIIIfx，45nm下跟费米大小差不多，性能大约有128Gflops，日本人还真是变态花这么高的代价 ...

兄弟，还是你一眼就看出成本的问题了；

舍得花钱还怕的不了第一的？天河只用了7千多个GPU，一个节点是2块CPU+1块GPU，一共只有7000多个1U的机箱的空间；

作者: 开普勒 时间: 2011-6-20 22:54
听说天-朝从今年开始也不用GPU搞了，为了自主创新，计划全部用龙芯搞。估计很快就会见到五位数龙芯堆出来的超算。

作者: luckissy 时间: 2011-6-20 22:56

asdfjkl 发表于 2011-6-20 22:46
兄弟，还是你一眼就看出成本的问题了；

舍得花钱还怕的不了第一的？天河只用了7千多个GPU，一个节点是 ...

{lol:]问题是**的龙芯花再多还是悲剧

作者: asdfjkl 时间: 2011-6-20 22:58
据富士通称，该款超级计算机仍在建造之中，将于2012年11月正式投入运行，届时处理器数量将增加到8万个。
怕是现在是第一；到了２０１１年１１月投入运行时，就不是第一了；呵呵呵，大家懂的！

作者: the_god_of_pig 时间: 2011-6-20 23:43
啥国产有什么意义

真需要我朝SNB一样造出来{lol:]

作者: SnakeLee 时间: 2011-6-21 03:17
日本人依旧顽固的思想，不过也挺好的...自己造舒心不是么~

作者: PaulWong 时间: 2011-6-21 08:27
山寨龙芯一出谁与争锋！

作者: nanshan 时间: 2011-6-21 08:40
没事，天河系统又可以名正言顺地申请国家拨款，重夺第一宝座，老黄又要笑眯眯地卖卡了

作者: nom8393 时间: 2011-6-21 08:44

pla16539544 发表于 2011-6-21 08:37
咦，没人说超算是堆出来的啦，果然东洋的月亮是比较圆的~如果是西洋的，估计还会更圆些~

日本这个京肯定是用“CPU堆出来”的，但是京相比天河一号有两个巨大的优势：1，京的CPU全部是日本自己的厂商富士通造的；
2，由于京没有使用GPU，那它的理论计算峰值是比较容易达到的。

PS：只要美帝不封锁，某大国由于不差钱，再堆出来N倍于现在天河一号的超级计算机根本不成问题。{:5_205:}

作者: nom8393 时间: 2011-6-21 10:11

pla16539544 发表于 2011-6-21 09:07
只要有钱，嘴一张就能堆个N倍超算出来~果然，天朝所以为天朝，人才在民间啊~赞叹一下

{:5_191:} 基本上就是这样啊，你不信看看最近有关超级计算机的新闻，其中一个就是中东某国家堆出来一个排名进TOP 50的超级计算机出来了。

PS：这不会打击到你的民族自信心和自豪感吧？

作者: whhwhhwhh 时间: 2011-6-21 10:31
前天在硅知识产权交易中心看到他们又在准备一个国产的自主产权32bit的CPU的发布会，是家苏州公司出的，哪个有料的来报下了。

作者: stefxia 时间: 2011-6-21 10:37
提示: 作者被禁止或删除内容自动屏蔽

作者: asdfjkl 时间: 2011-6-21 12:27

pla16539544 发表于 2011-6-21 08:37
咦，没人说超算是堆出来的啦，果然东洋的月亮是比较圆的~如果是西洋的，估计还会更圆些~

这大家都懂得呀，用了ＮＶ的就是堆出来的，浪费纳税人的钱了；要是用了ＡＭＤ，肯定没这话了！
说天河一号太强，没必要！　浪费是这帮子人！
如今天河一号被爆，说不行的；夸日本机器牛掰的；也是这帮子人！

作者: 左脚选手 时间: 2011-6-21 12:48

asdfjkl 发表于 2011-6-21 12:27
这大家都懂得呀，用了ＮＶ的就是堆出来的，浪费纳税人的钱了；要是用了ＡＭＤ，肯定没这话了！
说天河一 ...

这里没人说日本不是堆出来，天河一开始是用AMD的4870X2的，也有很多人说是堆出来的，效率不高，面子工程，难道你是选择性【请注意用词】？话说回来，日本自己造CPU，比我们天河要用intel cpu和NV的GPU那是当然牛的。

作者: nom8393 时间: 2011-6-21 13:23

asdfjkl 发表于 2011-6-21 12:27
这大家都懂得呀，用了ＮＶ的就是堆出来的，浪费纳税人的钱了；要是用了ＡＭＤ，肯定没这话了！
说天河一 ...

话可不能这么说，天河一号A卡和N卡都用了（一期是AMD的GPU，二期换了Telsa）。

作者: chinayyj 时间: 2011-6-21 14:18
這個排位是不是有問題！還沒建成的計算機都可以算嗎！或者在建造中途出了什麼意外怎辦？

作者: bird11530000 时间: 2011-6-21 14:54
用过4核的sparcv9，那速度简直想让我跑到美国把服务器砸了

作者: nom8393 时间: 2011-6-21 17:17

yysqu 发表于 2011-6-21 16:39
堆CPU就可以了？那还造什么高性能处理器，直接买5元一个P3堆不就得了。

问题是P3那运算能力，起码要堆百八十万块才能到目前主流的超级计算机水平（还不算系统总线的问题），你哪来这么多废P3？

作者: asdfjkl 时间: 2011-6-21 21:10

chinayyj 发表于 2011-6-21 14:18
這個排位是不是有問題！還沒建成的計算機都可以算嗎！或者在建造中途出了什麼意外怎辦？

嘿嘿嘿，你和我一样一开始开错了；文章是说8万颗的整体规模要到明年才建好；但是现在5万多颗CPU的超级计算机已经整好了。

作者: 美獣 时间: 2011-6-22 11:03
美国哪天不往外出售芯片了，超算性能版上能倒掉90%，日本这个堆CPU的反而还能保持第一

欢迎光临 POPPUR爱换 (https://we.poppur.com/)