2012年6月19日 星期二

supercomputers, High Performance Computing

Home | TOP500 Supercomputing Sites

www.top500.org/ - 頁庫存檔 - 翻譯這個網頁
In June 2011, the TOP500 ranked K the world's fastest supercomputer, with a rating of over 8 petaflops, and in November 2011, K became the first computer to ...


TOP500 List Releases ... The 39th TOP500 list will be released ...

November 2010

The 36th edition of the closely watched TOP500 list of the ...

TOP500 List - June 2011 (1-100)

Rank, Site, Computer/Year Vendor, Cores, Rmax, Rpeak, Power. 1 ...

November 2003

Search. Message-passing will become the minority way of ...

TOP500 List - November 2010 ...

Rank, Site, Computer/Year Vendor, Cores, Rmax, Rpeak, Power. 1 ...

November 2011

Japan's “K Computer” maintained its position atop the newest ...

20年前寫的超級電腦介紹 內容會大同小異

Computing From Weather to Warcraft

Published: November 17, 2008
For years, Western governments have used supercomputers to model weapons of nuclear war.

Now a company in China uses the powerful machines to tend the fantasy realms of World of Warcraft.
Supercomputers, which are up to a million times faster than the typical desktop PC, are still staples in the data warehouses of national laboratories and universities in the United States, Japan and Western Europe. But over the last few years, the falling cost of supercomputer systems has allowed a broader range of corporations and institutions, including many in China and India, to buy them for everything from processing movie graphics to searching for oil.
Just 18 months ago, China and India lacked a single system among the 25 fastest in the world. But on the latest list of the 500 fastest computers, released Monday, China nailed the No. 10 spot, standing as the only nation besides the United States in the top 10. India, meanwhile, had the 13th-fastest machine, beating Japan, a longtime leader.
China now claims 15 of the world’s 500 fastest computers. That makes it the top-ranking supercomputing country outside the United States, Western Europe and Japan.
The presence of supercomputers in emerging nations like China and India says as much about those countries’ growing national ambitions as the changing state of science and business.
“These other countries are following behind the U.S. and perhaps some other nations in Western Europe, but they are there,” said Jack Dongarra, a computer scientist at the University of Tennessee who helps maintain Top500, the official list of the fastest supercomputers. “These countries are making a clear statement about their intentions.”
The vast majority of supercomputers are built by I.B.M. and Hewlett-Packard. But China’s top system, located at the Shanghai Supercomputer Center, was assembled by the Chinese manufacturer Dawning. Like many of the fastest machines, the Shanghai system will handle research tasks, which remain the most important role for supercomputers. The ability of these machines to simulate experiments, explosions and the weather makes them crucial in an age when scientific discovery often takes place by manipulating large databases of information instead of running physical experiments.
“They are not buying these machines because they like to burn electricity and heat the air,” said Mark Seager, head of advanced computing at the Lawrence Livermore National Laboratory. “It’s for the simulation capabilities, which will be an important economic driver not just for the U.S., but for anyone else with two neurons to rub together.”
Still, the sharply falling cost of fast computers, which are often created by yoking together thousands of standard servers, makes them attractive to businesses for uses that would have been impractical even a few years ago.
For example, in 2007, the Tata Group, an Indian conglomerate, invested $35 million in a computing subsidiary that built what was then the fourth-fastest system in the world. Tata hopes to turn the machine into the basis of a profitable business with government contracts and work for researchers and companies in fields like nanotechnology, biology and electromagnetics. Tata’s computer is already being used to simulate aircraft designs for Boeing and render animated movies.
“We haven’t recovered our investment yet,” said Sunil Sherlekar, head of the Tata lab. “We don’t expect this to be hugely profitable in the short term, but we understand this is a long-term activity.”
For years, some of the fastest machines in China have belonged to The9, a video game developer that owns the local distribution rights to Blizzard Entertainment’s World of Warcraft franchise. Earlier this year, The9 boasted of hosting more than one million World of Warcraft players online at the same time. To support the complex calculations required to create the game’s graphics, The9 owns more than 10 supercomputer systems.
The list of China’s fastest computers is also filled with systems owned by oil and gas companies, financial firms, research groups and other media companies.
Of all the new entrants to the supercomputing race, China appears the most focused. The government has spent a vast amount of money building out its computing infrastructure, hoping to improve science and industry.
“If you look at China and what they are spending to get ahead, it’s clear this is a national priority,” said Douglas Comer, vice president of research at Cisco Systems. “They are definitely coming from behind, and they know that. They’re hungry.”
New Zealand is the leader in terms of computing capacity per capita, thanks to Weta Digital, a visual-effects company whose founders include the movie director Peter Jackson. Weta, based in the New Zealand capital, Wellington, operates four of the fastest machines on the planet for its work on film franchises like “The Lord of the Rings” and “The Fantastic Four.” Weta also rents out space on its systems to local research labs.
The idea of renting out space on big machines harks back to the early era of computing, when computers were so expensive that customers bought blocks of time on them for specific tasks. Today, a number of companies, like the high-speed computing specialist Cray and the graphics-chip maker Nvidia, are building beefy systems that can sit next to a desk and replicate some of the functions handled by room-sized machines. Nvidia, for example, has started selling deskside machines starting at $10,000 that can process data 250 times faster than a regular PC.
The goal behind such computers is to provide scientists, engineers and artists with direct access to strong machines before they send larger jobs off to supercomputers.
And, of course, there remains a prominent place for machines that can cost more than $100 million and take care of the United States government’s most secret jobs. The current fastest computer in the world is operated by Los Alamos National Laboratory in New Mexico, which uses it to perform classified military work.

IBM Breaks Performance Records Through Systems Innovation
PR-USA.net (press release) - Varna,Bulgaria
Engineers and researchers at the IBM ( NYSE : IBM) Hursley development lab in England and the Almaden Research Center in California have demonstrated ...

High Performance Computing 高效能運算設備

high-performance computing指作科學研究等的高速運算(High-speedcomputing, which typicallyrefers to supercomputers used in scientificresearch.)
1.適合對象:研究實驗內容需要利用電腦程式計算,且該程式之執行時間、記憶體利用等需求超過一般個人電腦。潛在使用者包括:a)自行使用C/C++/Fortran 等程式語言開發程式之研究團隊。 b) 目前已經透過 MPI OpenMP 等函式庫進行平行化程式撰寫之研究團隊。
2. 預期成效:將提供兩種不 同性質之高效能運算設備: a)SMP大型主機:具備256GB 記憶體,適合單一 程式需要大量記憶體的情況使用。此外對於已經利用OpenMP進行平行化運算的 程式,也可以提供有效幫助。 b)cluster 叢集伺服器,提供超過一百組節點之 運算主機,系統運算能力高達 3TFlops,特別適合已經利用MPI函式庫進行平行化之程式。 c)教育訓練:未來將與廠商及校內教授合作,提供多元化教育訓練予校內研究團隊,期能幫助校內研究團隊熟悉各式主機之使用、程式撰寫與除錯技巧、平行化程式之設計與開發,以便有效協助研究團隊,加速各式實驗之進行。
大型的網路公司如古鉤(Google)、電子灣(eBay)及亞馬遜擁有的電腦運算力量,比起任何學術電腦網都要大,而且目前已成為配銷大規模電腦運算服 務的領頭者。這些公司的許多創新技術都可以幫忙科學家,例如今年八月推出的「點播運算」(computing-on-demand)服務正是其一,客戶使 用亞馬遜龐大的電算基礎設施,創設虛擬多重的電腦,每一運算小時只要美金一角;另外,每十億位元組(gigabyte)資訊的儲存費用,每月只要美金一角 五分。......

這 一點很了不起。在單一伺服器上運作多重的虛擬電腦,使用資源時有效率得多。此外,這一點還意味著不必安裝有實體、擁有特殊操作系統的機器,只消幾秒鐘, 就能造出虛擬版的電腦來。如此的虛擬電腦可以像檔案般複製,還可以在任何機器上來跑,不管機器用的硬體是什麼。佩斯表示,以往大家每台機器都得裝軟、硬 體,現在則不一定了。
要取得虛擬化軟體,目前愈來愈形便利,比如微軟、Vmware等公司都有提供。亞馬遜使用的軟體叫「仙」(Xen), 這個由英國劍橋大學研發出來、程式碼 開放的系統,很快就在學者間流行起來。學者可以使用「仙」這套系統,可以讓虛擬電腦來操作一整個網格或一整叢的電腦(儘管每台電腦的操作系統都不相同), 還可以使用自己實驗室裡開發出來的應用程式。
--Knowledge Review 知識新知 12/16/2006


Intel has overhauled the basic building block of the information age, paving the way for faster and more energy-efficient processors.

---- To me the attempt to redefine it as "High Productivity Computing" is, if I may be to the point, nothing more than a group of people attempting to justify the (probably overpriced) services they are now attempting to sell by a meaningless (from a practical perspective) redefinition of an acronym. I'm kinda surprised that they're not pushing SSHPC (Six Sigma...).

IBM Roadrunner, a supercomputer
Wikipedia article "Geococcyx".Road running is the sport of running on a measured course over an established road (as opposed to track and cross country running). These events normally range from 5 km to long distance, such as half marathons and marathons, and may involve large numbers of runners or wheelchair entrants. Road running is part of group of road events known as "road races".

Military Supercomputer Sets Record

Published: June 9, 2008

SAN FRANCISCO — An American military supercomputer, assembled from components originally designed for video game machines, has reached a long-sought-after computing milestone by processing more than 1.026 quadrillion calculations per second.
Skip to next paragraph

The Roadrunner supercomputer costs $133 million and will be used to study nuclear weapons.

The new machine is more than twice as fast as the previous fastest supercomputer, the I.B.M. BlueGene/L, which is based at Lawrence Livermore National Laboratory in California.
The new $133 million supercomputer, called Roadrunner in a reference to the state bird of New Mexico, was devised and built by engineers and scientists at I.B.M. and Los Alamos National Laboratory, based in Los Alamos, N.M. It will be used principally to solve classified military problems to ensure that the nation’s stockpile of nuclear weapons will continue to work correctly as they age. The Roadrunner will simulate the behavior of the weapons in the first fraction of a second during an explosion.
Before it is placed in a classified environment, it will also be used to explore scientific problems like climate change. The greater speed of the Roadrunner will make it possible for scientists to test global climate models with higher accuracy.
To put the performance of the machine in perspective, Thomas P. D’Agostino, the administrator of the National Nuclear Security Administration, said that if all six billion people on earth used hand calculators and performed calculations 24 hours a day and seven days a week, it would take them 46 years to do what the Roadrunner can in one day.
The machine is an unusual blend of chips used in consumer products and advanced parallel computing technologies. The lessons that computer scientists learn by making it calculate even faster are seen as essential to the future of both personal and mobile consumer computing.
The high-performance computing goal, known as a petaflop — one thousand trillion calculations per second — has long been viewed as a crucial milestone by military, technical and scientific organizations in the United States, as well as a growing group including Japan, China and the European Union. All view supercomputing technology as a symbol of national economic competitiveness.
By running programs that find a solution in hours or even less time — compared with as long as three months on older generations of computers — petaflop machines like Roadrunner have the potential to fundamentally alter science and engineering, supercomputer experts say. Researchers can ask questions and receive answers virtually interactively and can perform experiments that would previously have been impractical.
“This is equivalent to the four-minute mile of supercomputing,” said Jack Dongarra, a computer scientist at the University of Tennessee who for several decades has tracked the performance of the fastest computers.
Each new supercomputing generation has brought scientists a step closer to faithfully simulating physical reality. It has also produced software and hardware technologies that have rapidly spilled out into the rest of the computer industry for consumer and business products.
Technology is flowing in the opposite direction as well. Consumer-oriented computing began dominating research and development spending on technology shortly after the cold war ended in the late 1980s, and that trend is evident in the design of the world’s fastest computers.
The Roadrunner is based on a radical design that includes 12,960 chips that are an improved version of an I.B.M. Cell microprocessor, a parallel processing chip originally created for Sony’s PlayStation 3 video-game machine. The Sony chips are used as accelerators, or turbochargers, for portions of calculations.
The Roadrunner also includes a smaller number of more conventional Opteron processors, made by Advanced Micro Devices, which are already widely used in corporate servers.
“Roadrunner tells us about what will happen in the next decade,” said Horst Simon, associate laboratory director for computer science at the Lawrence Berkeley National Laboratory. “Technology is coming from the consumer electronics market and the innovation is happening first in terms of cellphones and embedded electronics.”
The innovations flowing from this generation of high-speed computers will most likely result from the way computer scientists manage the complexity of the system’s hardware.
Roadrunner, which consumes roughly three megawatts of power, or about the power required by a large suburban shopping center, requires three separate programming tools because it has three types of processors. Programmers have to figure out how to keep all of the 116,640 processor cores in the machine occupied simultaneously in order for it to run effectively.
“We’ve proved some skeptics wrong,” said Michael R. Anastasio, a physicist who is director of the Los Alamos National Laboratory. “This gives us a window into a whole new way of computing. We can look at phenomena we have never seen before.”
Solving that programming problem is important because in just a few years personal computers will have microprocessor chips with dozens or even hundreds of processor cores. The industry is now hunting for new techniques for making use of the new computing power. Some experts, however, are skeptical that the most powerful supercomputers will provide useful examples.
“If Chevy wins the Daytona 500, they try to convince you the Chevy Malibu you’re driving will benefit from this,” said Steve Wallach, a supercomputer designer who is chief scientist of Convey Computer, a start-up firm based in Richardson, Tex.
Those who work with weapons might not have much to offer the video gamers of the world, he suggested.
Many executives and scientists see Roadrunner as an example of the resurgence of the United States in supercomputing.
Although American companies had dominated the field since its inception in the 1960s, in 2002 the Japanese Earth Simulator briefly claimed the title of the world’s fastest by executing more than 35 trillion mathematical calculations per second. Two years later, a supercomputer created by I.B.M. reclaimed the speed record for the United States. The Japanese challenge, however, led Congress and the Bush administration to reinvest in high-performance computing.
“It’s a sign that we are maintaining our position,“ said Peter J. Ungaro, chief executive of Cray, a maker of supercomputers. He noted, however, that “the real competitiveness is based on the discoveries that are based on the machines.”
Having surpassed the petaflop barrier, I.B.M. is already looking toward the next generation of supercomputing. “You do these record-setting things because you know that in the end we will push on to the next generation and the one who is there first will be the leader,” said Nicholas M. Donofrio, an I.B.M. executive vice president.
By breaking the petaflop barrier sooner than had been generally expected, the United States’ supercomputer industry has been able to sustain a pace of continuous performance increases, improving a thousandfold in processing power in 11 years. The next thousandfold goal is the exaflop, which is a quintillion calculations per second, followed by the zettaflop, the yottaflop and the xeraflop.

 Advanced Research Computing @ Cardiff (ARCCA),


Making the impossible possible

2 June 2008
Information Services director Martyn Harrow (left) and ARCCA director Professor Martyn Guest with the new Bull High Performance Computer
Information Services director Martyn Harrow (left) and ARCCA director Professor Martyn Guest with the new Bull High Performance Computer
Academics across the University are set to benefit after the launch of a powerful new computing cluster which will enable research projects previously considered too difficult or time-consuming.
Supplied by leading international IT firm Bull, the new computer is one of the most advanced in the UK academic sector.
The computer will be run by Advanced Research Computing @ Cardiff (ARCCA), which was set up to supply all University academics with the high-powered technology necessary to tackle today’s big research questions. Already ARCCA is putting its computing power to work in a wide variety of fields. These include:
  • Health. Working with the new Positron Emission Tomography scanner (PET), able to detect cancers at a smaller size than previous technology. A separate project involves the School of Computer Science and Velindre Cancer Centre in developing more accurate radiotherapy plans for cancer patients.
  • Neuroscience. Working with the Cardiff University Brain Research Imaging Centre to map the structure and function of our brains.
  • Geosciences. Simulating earth mantle and tectonic plate movements to improve our understanding of earthquakes and volcano eruptions.
  • Astrophysics. Recreating the formation of stars and planets, and taking part in the international hunt for gravitational waves.
  • Archaeology. Working with English Heritage to pinpoint the carbon dating of prehistoric sites.
  • Renewable Energy. Working with engineers to model hydrodynamic processes which can be used for tidal and wave power.
The High Performance Computer will support study in all areas of research, including the arts, humanities and social sciences. Cardiff Business School is already working with ARCCA on economic modelling, and the School of English, Communication and Philosophy on linguistics.
The computer uses Intel® Xeon® Quad-core processors, with four cores to each chip. The system has approximately four terabytes (or four million megabytes) of memory and has just been measured as performing 20 trillion floating point operations a second (20 Teraflops). These results have yet to be officially ratified but would make it the most powerful cluster in a UK University dedicated to in-house research. It was funded with a Science Research Investment Funding (SRIF) grant from the Higher Education Funding Council Wales
The computer will not only be one of the most powerful at a British University, but also one of the greenest. Based in its own state-of-the-art data centre, it is housed in ten energy efficient water-cooled racks, saving around £30,000 a year on conventional air cooling systems.
Cardiff’s partnership with Bull will continue with the creation of the Cardiff High Performance Computing Centre of Excellence, based around the new computer. The Centre will extend the scope and quality of computer-based research support and open up a range of new research frontiers.
Launching the High Performance Computer at the University, Welsh Assembly Government First Minister, Rhodri Morgan said: “The developments in High Performance Computing brought about by ARCCA are already making huge differences in many areas of research. This puts Cardiff University at the forefront of computer-based research techniques in Wales and the UK, as well as internationally.”
Bull High Performance Computer
Didier Lamouche, Bull Chief Executive Officer, said: “Being involved in this partnership with Cardiff University has enabled us to demonstrate the importance of utilising leading-edge IT resources to pursue ground-breaking research. This new supercomputer will support Cardiff’s growing reputation as one of the most innovative, ambitious and successful universities in the country and internationally.”
The Vice-Chancellor of Cardiff University, Dr David Grant, said: “The technical specifications of the Bull High Performance Computer are extremely impressive. We expect the research enabled by this computational power will be more impressive still. Computer modelling is becoming vital to our understanding of human biology and the development of new drugs. Simulation will bring major benefits in the sciences and engineering, and open up completely new research fields in the arts, humanities and social sciences. The new Cardiff High Performance Computing Centre of Excellence will keep the University at the forefront of these exciting possibilities in all of its academic disciplines.”
Professor Martyn Guest, Director of ARCCA, said: “ARCCA exists to deliver high-end computing service, resources and support to all researchers in all disciplines and Schools. The new High Performance Computer will allow a wide variety of studies previously dismissed as impossible or impractical. We look forward to talking to academics from all fields about how ARCCA can help them achieve their research objectives.”

Related links


卡迪夫大學新置的超級計算機計算速度可達每秒20 teraflops
英國加的夫大學(Cardiff University
價值290萬英鎊的這台高性能計算機是法國布爾(Bull)集團提供的,計算能力可達每秒鐘20萬億次浮點運算(20 teraflops)。
加的夫大學是英國研究能力較強的學府之一,現有兩名諾貝爾獎得主:其一是獲得2007年醫學獎的馬丁•埃文思教授(Prof Martin Evans),另一位是1988年化學獎得主羅伯特•休伯爾教授(Prof Robert Huber)。

在醫學領域,它將能協助正電子斷層掃描儀(Positron Emission Tomography-PET)檢測出更小範圍的癌病變。
這台超級計算機也會用於考古研究。目前該校正與英格蘭文化遺產協會(English Heritage)合作,展開對遠古遺址年代的炭鑑定。
加的夫大學校長戴維•格蘭特(David Grant)說,計算機模型在當今的生理學研究和新藥開發方面已在扮演必不可少的角色。

卡迪夫大学新置的超级计算机计算速度可达每秒20 teraflops
英国加的夫大学(Cardiff University

价值290万英镑的这台高性能计算机是法国布尔(Bull)集团提供的,计算能力可达每秒钟20万亿次浮点运算(20 teraflops)。
加的夫大学是英国研究能力较强的学府之一,现有两名诺贝尔奖得主:其一是获得2007年医学奖的马丁•埃文思教授(Prof Martin Evans),另一位是1988年化学奖得主罗伯特•休伯尔教授(Prof Robert Huber)。
  • 在医学领域,它将能协助正电子断层扫描仪(Positron Emission Tomography-PET)检测出更小范围的癌病变。
  • 计算机也将用于该校正与威尔士韦林德尔(Velindre)肿瘤中心一起研究的放射治疗照射量的计算,以便给病人提供量度更精确的照射治疗。
  • 在神经科学领域,加的夫大学的人脑研究扫描中心可望使用这台计算机绘制人脑结构与功能图。
  • 这台超级计算机也会用于考古研究。目前该校正与英格兰文化遗产协会(English Heritage)合作,展开对远古遗址年代的炭鉴定。
  • 可再生能源的研究也是这台电脑效力的领域,譬如把它用于制作水利动力过程的模型。这类模型也可用于开发洪水预报的技术。
加的夫大学校长戴维•格兰特(David Grant)说,计算机模型在当今的生理学研究和新药开发方面已在扮演必不可少的角色。