Sunday, June 22, 2008

Roadrunner is the Fastest Supercomputer

Roadrunner Roadrunner (photo courtesy: Los Alamos National Laboratory, Berkeley)

With the publication of the latest edition of the TOP500 list of the world’s most powerful supercomputers, the global high performance computing community has officially entered a new realm—a supercomputer with a peak performance of more than 1 petaflop/s (one quadrillion floating point operations per second).

The new No. 1 system, built by IBM for the U.S. Department of Energy’s Los Alamos National Laboratory and and named “Roadrunner,” by LANL after the state bird of New Mexico achieved performance of 1.026 petaflop/s—becoming the first supercomputer ever to reach this milestone. At the same time, Roadrunner is also one of the most energy efficient systems on the TOP500.

The 31st edition of the TOP500 list was released at the International Supercomputing Conference in Dresden, Germany. Since 1993, the list has been produced twice a year and is the most extensive survey of trends and changes in the global supercomputing arena.

The Roadrunner system is based on the IBM QS22 blades which are built with advanced versions of the processor in the Sony PlayStation 3, displaces the reigning IBM BlueGene/L system at DOE’s Lawrence Livermore National Laboratory. Blue Gene/L, with a performance of 478.2 teraflop/s (trillions of floating point operations per second) is now ranked No. 2 after holding the top position since November 2004.

Rounding out the top five positions, all of which are in the U.S., are the new IBM BlueGene/P (450.3 teraflop/s) at DOE’s Argonne National Laboratory, the new Sun SunBlade x6420 “Ranger” system (326 teraflop/s) at the Texas Advanced Computing Center at the University of Texas – Austin, and the upgraded Cray XT4 “Jaguar” (205 teraflop/s) at DOE’s Oak Ridge National Laboratory.

Among all systems, Intel continues to power an increasing number, with Intel processors now found in 75% of the TOP500 supercomputers, up from 70.8% of the 30th list released last November.

Other highlights from the latest list include:
-- Quad-core processor based systems have taken over the TOP500 quite rapidly. Already 283 systems are using them. Two hundred three systems are using dual-core processors, only eleven systems still use single core processors, and three systems use IBMs advanced Sony PlayStation 3 processor with 9 cores.
-- The top industrial customer, at No. 10, is the French oil company: Total Exploration Production.
-- IBM held on to its lead in systems with 210 systems (42%) over Hewlett Packard with 183 systems (36.6%). IBM had 232 systems (46.4%) six months ago, compared to HP with 166 systems (33.2%).
-- IBM remains the clear leader in the TOP500 list in performance with 48% of installed total performance (up from 45), compared to HP with 22.4% (down from 23.9). In the system category Dell, SGI and Cray follow with 5.4%, 4.4% and 3.2% respectively.
-- The last system on the list would have been listed at position 200 in the previous TOP500 just six months ago. This is the largest turnover rate in the 16-year history of the TOP500 project.

For the first time, the TOP500 list will also provide energy efficiency calculations for many of the computing systems and will continue tracking them in consistent manner. Most energy efficient supercomputers are based on
IBM QS22 Cell processor blades (up to 488 Mflop/s/Watt),
IBM BlueGene/P systems (up to 376 Mflop/s/Watt)
Intel Harpertown quad-core blades are catching up fast:
IBM BladeCenter HS21with low-power processors (up to 265 Mflop/s/Watt)
SGI Altix ICE 8200EX Xeon quad-core nodes, (up to 240 Mflop/s/Watt) ,
Hewlett-Packard Cluster Platform 3000 BL2x220 with double density blades (up to 227 Mflop/s/Watt).

The U.S. is clearly the leading consumer of HPC systems with 257 of the 500 systems. The European share (184 systems – up from 149) is still rising and is again larger then the Asian share (48 – down from 58 systems). Dominant countries in Asia are Japan with 22 systems (up from 20), China with 12 systems (up from 10), India with 6 systems (down from 9), and Taiwan with 3 (down from 11).

Wednesday, June 11, 2008

Robert Langer Awarded 2008 Millennium Technology Prize for Intelligent Drug Delivery

Robert Langer Prof. Robert langer [photo courtesy: Stu Rosner Photograhy & Department of Chemical Engineering, MIT]

Today at Helsinki, the 2008 Millennium Technology Prize was presented to Professor Robert Langer of Massachusettes Institute of Technology and Harvard-MIT Division of Health Sciences and Technology for developing innovative biomaterials for controlled drug release.

The world’s largest technology prize is awarded by Technology Academy Finland for a technological innovation that significantly improves quality of human life and promotes sustainable development. President of the Republic of Finland Tarja Halonen handed Professor Langer the prize of EUR 800,000 and ”Peak”, the prize trophy, at the Grand Award Ceremony on Wednesday afternoon.

Professor Robert Langer's innovations have had a significant impact on fighting cancer, heart disease, and numerous other diseases. His work has also brought about significant advances in tissue engineering, including synthetic replacement for biological tissues such as artificial skin. Over 100 million people a year are already using advanced drug delivery systems and this number is rising rapidly. In the future, tissue engineering may revolutionize medical treatment that could affect millions of other individuals.

"Tissue engineering holds the promise of creating virtually any new tissue or organ," said Professor Langer.Known as the father of controlled drug delivery and tissue engineering, Professor Langer has been cited as "one of history's most prolific inventors in medicine". Professor Langer's research laboratory at MIT is the largest biomedical engineering laboratory in the world.

The Millennium Technology Prize is awarded every second year. The 2004 award was given to Tim Berners-Lee who revolutionized electronic communications with the invention of the World Wide Web. The 2006 award was given to Shuji Nakamura, a professor at the University of California at Santa Barbara and inventor of the high-brightness gallium nitride light-emitting diode (LED) and a blue laser.

Please read our past posting dated April 8, 2008 to know about other contender's of this year's award.

Wednesday, April 23, 2008

IBM Unveils High-powered 'iDataPlex' Server for Web 2.0 Computing

iDataPlexiDataPlex uses 40 percent less power while increasing the amount of computing that can be done 5X. To achieve these breakthroughs IBM created a design that, among other things, turns the standard rack on its side [Image courtesy: IBM]

Today, IBM unveiled a new type of server design that targets so-called Web 2.0 companies operating or building massive data centers with tens of thousands of servers to power social networks, online gaming, search, and other services that reach millions of users.

The iDataPlex system, which leverages IBM's Intel-based blade-server technology, doubles the number of servers that can run in a vertical rack, while using 40% less power. The new system uses a liquid-cooled wall in the back to provide cooling, which enables the technology to run without the need of air conditioning.

The rack system is meant to replace the white-box servers commonly found in Internet companies' data centers. The iDataPlex is designed for "stateless computing" that turns many separate computers into a pool of shared resources, IBM said. The system is geared toward allowing customers to scale rapidly, as traffic increases.

"With iDataPlex, IBM is making Web 2.0-style computing more efficient and commercializing it for Internet companies and other high performance segments like financial services and research," Bill Zeitler, senior VP of IBM's systems and technology group, said in a statement. "iDataPlex can provide a foundation that companies can build on to provide improved services to Web users around the world."

The system runs the open-source Linux operating system, including distributions from IBM partners Novell and Red Hat. It also uses the open source, scale-out cluster management software xCat. IBM plans to use iDataPlex in its Cloud Computing Center at Dublin, Ireland and at the Almaden Research Center in San Jose, Calif.

Vendors that have committed to work with IBM in building products for iDataPlex include Avocent, Blade Network Technology, Devon IT, Force 10 Networks, Intel, QLogic, and SMC Networks. Blade Network, for example, built its new RackSwitch G8000 Ethernet switch specifically for the new system, IBM said.

IDataPlex is scheduled for release in the United States and Canada next month and globally by the end of the year. The new product is a custom-built solution, so pricing, which could run into the millions of dollars, will depend on configuration.

Wednesday, April 16, 2008

Spintronics based 'Racetrack' Memory from IBM

IBM LogoIn two papers published in the April 11 issue of Science, IBM Fellow Stuart Parkin and colleagues at the IBM Almaden Research Center in San Jose describe both the fundamentals of a next-generation nonvolatile memory dubbed "racetrack" as well as a milestone in that technology. "Racetrack" is expected to initially replace flash memory and eventually hard-disk drives.
Using spintronics --the storage of bits generated by the magnetic spin of electrons rather than their charge--a proof-of-concept shift register was demonstrated by this team from IBM. The prototype encodes bits into the magnetic domain walls along the length of a silicon nanowire, or 'Racetrack', so named because the data "races" around the wire "track". IBM uses "massless motion" to move the magnetic domain walls along the nanowire for the storage and retrieval of information.

With these advancements, computer memory that combines the high performance and reliability of flash with the low cost and high capacity of the hard disk drive could be closer than we dreamed of. This milestone could lead to electronic devices capable of storing far more data in the same amount of space than is possible today, with lightning-fast boot times, far lower cost and unprecedented stability and durability.

Currently, there are two main ways to store digital information: solid state random access flash memory, commonly used in devices such as mobile phones, music players and digital cameras, and the magnetic hard disk drive, commonly used in desktop and laptop computers and some handheld devices.

While both classes of storage devices are evolving at a very rapid pace, the cost of storing a single data bit in a hard disk drive remains approximately 100 times cheaper than in flash memory. While the low cost of the hard disk drive is very attractive, these devices are intrinsically slower and, with many moving parts, have mechanical reliability issues not present in flash technologies.
Flash memory, however, has its own drawbacks – while it is fast to read data, it is slow to write data, and it, too, has a finite lifespan. Flash, can be reused only a few thousands of times because it eventually breaks because it is slightly damaged by each use or "rewrite."

Within the next ten years, racetrack memory could lead to solid state electronic devices – with no moving parts, and therefore more durable – capable of holding far more data in the same amount of space than is possible today. For example, this could enable a handheld device such as an mp3 player to store around 500,000 songs or around 3,500 movies – 100 times more than is possible today – with far lower cost and power consumption. The devices would not only store vastly more information in the same space, but also require much less power and generate much less heat, and be practically unbreakable; the result: massive amounts of personal storage that could run on a single battery for weeks at a time and last for decades.

"Current-Controlled Magnetic Domain-Wall Nanowire Shift Register"

Masamitsu Hayashi, Luc Thomas, Rai Moriya, Charles Rettner, Stuart S. P. Parkin
Science, Vol. 320. no. 5873, pp. 209 - 211 (11 April, 2008). Abstract Link
[2] "Magnetic Domain-Wall Racetrack Memory"
Stuart S. P. Parkin, Masamitsu Hayashi, Luc Thomas
Science, Vol. 320. no. 5873, pp. 190 - 194 (April 11, 2008). Abstract Link