super computer by shreyashukla

VIEWS: 810 PAGES: 63

									-----------------------------------------------------------------------

Supercomputer
-----------------------------------------------------------------------

Presented by: guided by: ----------------------------------------------------------------Shreya shukla Ayesha sheikh

Hasmukh goswami college of engg.

Supercomputers
Overview Supercomputers can be defined as the most advanced and powerful computers, or array of computers, in existence at the time of their construction. Supercomputers are used to solve problems that are too complex or too massive for standard computers, like calculating how individual molecules move in a tornado, or forecasting detailed weather patterns. Some supercomputers are single computers consisting of multiple processors; others are clusters of computers that work together. History Supercomputers were first developed in the early 1970s when Seymour Cray introduced the “Cray 1” supercomputer. Because microprocessors were not yet available, the processor consisted of individual integrated circuits. Successive generations of supercomputers were developed by Cray and became more powerful with each version. Other companies like IBM, NEC, Texas Instruments and Unisys began to design and manufacture more powerful and faster computers after the introduction of the Cray 1. You can read more about Seymour Cray and other leading figures in supercomputer technology at www.computerhalloffame.org/. The history of supercomputers can be viewed in detail at this Web page. Visit the Digital Century Web site to find a general overview of the history of the development of computers. Today's fastest supercomputers include IBM's Blue Gene and ASCI Purple, SCC's Beowulf, and Cray's SV2. These supercomputers are usually designed to carry out specific tasks. For example, IBM's ASCI Purple is a $250 million supercomputer built for the Department of Energy (DOE). This computer, with a peak speed of 467 teraflops, is used to simulate aging and the operation of nuclear weapons. Learn all about this project by linking to this article. Future supercomputer designs might incorporate the use of entirely new technologies of circuit miniaturization that could include new storage devices and data transfer systems. Scientists at UCLA are currently working on computer processor and circuit designs involving a series of molecules that behave like transistors. By incorporating this technology, new designs might include processors 10,000 times smaller, yet much more powerful than any current models. A comprehensive article about this research can be found at www.applesforhealth.com/supercomp1.html. Processing Speeds Supercomputer computational power is rated in FLOPS (Floating Point Operations Per Second). The first commercially available supercomputers reached speeds of 10 to 100 million FLOPS. The next generation of supercomputers (some of which are presently in the early stages of development) is predicted to break the petaflop level. This would represent computing power more than 1,000 times faster than a teraflop machine. To put these processing speeds in perspective, a

relatively old supercomputer such as the Cray C90 (built in the mid to late 1990s) has a processing speed of only 8 gigaflops. It can solve a problem, which takes a personal computer a few hours, in .002 seconds! The site www.top500.org/ is dedicated to providing information about the current 500 sites with the fastest supercomputers. Both the list and the content at this site is updated regularly, providing those interested with a wealth of information about the developments in supercomputing technology. Supercomputer Architecture Supercomputer design varies from model to model. Generally, there are vector computers and parallel computers. Detailed information about both kinds of architecture can be found at www.sdsc.edu/discovery/lo/sc.htm. Vector computers use a very fast data “pipeline” to move data from components and memory in the computer to a central processor. Parallel computers use multiple processors, each with their own memory banks, to 'split up' data intensive tasks. A good analogy to contrast vector and parallel computers is that a vector computer could be represented as a single person solving a series of 20 math problems in consecutive order; while a parallel computer could be represented as 20 people, each solving one math problem in the series. Even if the single person (vector) were a master mathematician, 20 people would be able to finish the series much quicker. Other major differences between vector and parallel processors include how data is handled and how each machine allocates memory. A vector machine is usually a single super-fast processor with all the computer's memory allocated to its operation. A parallel machine has multiple processors, each with its own memory. Vector machines are easier to program, while parallel machines, with data from multiple processors (in some cases greater than 10,000 processors), can be tricky to orchestrate. To continue the analogy, 20 people working together (parallel) could have trouble with communication of data between them, whereas a single person (vector) would entirely avoid these communication complexities. Recently, parallel vector computers have been developed to take advantage of both designs. For more information about this design, visit this Netlib.org page. Uses of Supercomputers Supercomputers are called upon to perform the most computeintensive tasks of modern times. As supercomputers have developed in the last 30 years, so have the tasks they typically perform. Modeling of real world complex systems such as fluid dynamics, weather patterns, seismic activity prediction, and nuclear explosion dynamics represent the most modern adaptations of supercomputers. Other tasks include human genome sequencing, credit card transaction processing, and the design and testing of modern aircraft. Manufacturers Although there are numerous companies that manufacture supercomputers, information about purchasing one is not always easy to find on the Internet. The price tag for a custom-built supercomputer can range anywhere from about

$500,000 for a beowulf system, up to millions of dollars for the newest and fastest supercomputers. Cray provides an informative Web site (www.cray.com/) with product descriptions, photos, company information, and an index of current developments. Scyld Computing Corporation (SCC) provides a Web site (www.scyld.com/) with detailed information about their Beowulf Operating System and the computers developed to allow multiple systems to operate under one platform. IBM has produced, and continues to produce, some of the most cutting-edge supercomputer technology. For information about IBM supercomputers visit www.ibm.com/. Their “Blue Gene” supercomputer, being constructed in collaboration with Lawrence Livermore National Labs, is expected to run 15 times faster (at 200 teraflops) than their current supercomputers. Read all about this project by visiting this link. IBM is also currently working on what they call a "self-aware" supercomputer, named "Blue Sky", for The National Center for Atmospheric Research (NCAR) in Boulder, Colorado. The Blue Sky will be used to work on colossal computing problems such as weather prediction. Additionally, this supercomputer can self-repair, requiring no human intervention. Read all about Blue Sky in the article found here. Intel has developed a line of supercomputers known as Intel TFLOPS. Supercomputers that use thousands of Pentium Pro processors in a parallel configuration to meet the supercomputing demands of their customers. Information about Intel supercomputers can be found at Intel's Web site (www.intel.com) or by reading this article.

What is a Supercomputer?
A supercomputer is defined simply as the most powerful class of computers at any point in time. Supercomputers are used to solve large and complex problems that are insurmountable by smaller, less powerful computers. Since the pioneering Cray-1® system arrived in 1976, supercomputers have made a significant contribution to the advancement of knowledge and the quality of human life. Problems of major economic, scientific and strategic importance typically are addressed by supercomputers years before becoming tractable on less-capable systems. In conjunction with some of the world's most creative scientific and engineering minds, these formidable tools already have made automobiles safer and more fuel-efficient; located new deposits of oil and gas; saved lives and property by predicting severe storms; created new materials and life-saving drugs; powered advances in electronics and visualization; safeguarded national security; and unraveled mysteries ranging from protein-folding mechanisms to the shape of the universe. Capable supercomputers are in short supply. Today's supercomputer market is replete with "commodity clusters," products assembled from collections of servers or PCs. Clusters are adept at tackling small problems and large problems lacking complexity, but are inefficient at the most demanding, consequential challenges - especially those of industry. Climate research algorithms, for example, are unable to achieve high levels of performance on these computers. The primary "design points" for today's clusters are server and PC markets, not supercomputing. Christopher Lazou, a high-performance computing consultant, explains, "Using tens of thousands of commodity chips may provide the capacity (peak flop rates) but not the capability, because of lack of memory bandwidth to a very large shared memory." Cray's product portfolio addresses this issue with high-bandwidth offerings. High-end Supercomputers

For important classes of applications, there is no substitute for supercomputers specifically designed not only for performance, but also high-bandwidth and low latency. Historically, this has been accomplished through vector architectures and more recently, multi-threaded architectures. These specialized supercomputers are built to meet the most challenging computing problems in the world. Today, new technology and innovation at Cray Inc. has allowed for a new class of supercomputers that combines the performance characteristics of vector supercomputers with the scalability of commodity clusters to achieve both high-efficiency and extreme performance in a scalable system architecture. These characteristics are embodied in the recently announced Cray X1™ system. The Future of Supercomputing Applications promising future competitive and scientific advantage create an insatiable demand for more supercomputer power - 10 to 1,000 times greater than anything available today, according to users. Automotive companies are targeting increased passenger cabin comfort, improved safety and handling. Aerospace firms envision more efficient planes and space vehicles. The petroleum industry wants to "see" subsurface phenomena in greater detail. Urban planners hope to ease traffic congestion. Integrated digital imaging and virtual surgery - including simulated sense of touch - are high on the wish list in medicine. The sequencing of the human genome promises to open an era of burgeoning research and commercial enterprise in the life sciences. As the demand for supercomputing power increases and the market expands, Cray's focus remains on providing superior real-world performance. Today's "theoretical peak performance" and benchmark tests are evolving to match the requirements of science and industry, and Cray supercomputing systems will provide the tools they need to solve their most complex computational problems

American Super Computer released to Russia a business adventure of ROY International Consultancy, Inc.!

May 18, 2000 - Moscow - The first Mainframe High Power Super Computer was exported from USA to Russia. This deal is a part of a long-term contract between the Russian Oil exploration firm Tatneftgeophysika and ROY International Consultancy Inc., headed by Dr.Cherian Eapen, who is shuttling between USA and Russia. Last Christmas day, the Super Computer - Sun Microsystem's "Starfire Enterprise10000" - was installed at the specially prepared site of the client in Bugulma, an interior town of Tatarstan Republic of Russia. President of Sun Microsystems Corporation, Mr. Scott McNeally, (who challenged Mr. Bill Gates and Microsoft for its technology and legal competency), congratulated the great effort of the President of ROY International Dr. Cherian Eapen, who is shuttling between USA and Russia and made the impossible a possible one. "It was a 'Christmas Starfire', a precious Christmas gift to Russia from America. This is an opening of high power computer usage in this geography for peaceful purposes - a new bridge opened between the two technology Super Powers of the world" - he said. The Starfire Enterprise10000 is purchased for the seismological data processing center of Tatneftgeophysika, a giant in the geophysical business field in the whole region of the former Soviet Union. In spite of the existing financial and economical problems, the Russian Geophysical firms are struggling hard to stand on their own legs by procuring the most modern technology stream in the field of computerization and processing of geological and geophysical data. By 1999, the majority of geophysical firms of Russia achieved the automation of their production centers. Year 2000 opens with the

2nd phase of modernization and reconstruction of geophysical computing centers, focusing mainly on upgrading of the power and speed of data interpretation. At present, the Russian seismological survey of Oil and Gas determines all their data based on 2 Dimensional film which should be obtained with 3 Dimensional (3D) film. Without 3D film, it is impossible for accurate and quality identification of Hydro-carbide fields. This 3D procedure increases the cost of complicated research work, with high power computers and software. But this gives a substantial economical advantage to the tune of 20 to 30%. In order to become competitive in the already saturated 3D seismic market, traditional geophysical firms started spending large amounts to modernize their computing centers. They started inviting companies specialized in this sphere -- Systems Integrators -- with given criterion of price, optimum efficiency, productivity of technical solution, taking into consideration all aspects of technology for processing of geophysical information. One such experienced Systems Integrators working in CIS is ROY International Consultancy Inc., whose main activity is the project design and realization of corporate computing, especially for computing centers for the oil and gas field. Founded in 1988, ROY International is the leading Systems Integrator specialized in large computer systems development of corporate computing centers. ROY International is the largest supplier of highly reliable and secure UNIX based enterprise wide systems in the CIS. By this period, ROY International designed and installed 300 projects throughout CIS countries to modernize and reconstruct computing centers, installing more than 2000 high power Sun Work Stations and Servers and porting major software, available in the world and networking etc. Bashkiristan Neftgeophysika, Udmrtneftgeophysika, Khantimansisk Geophysika, Sakhalinsk Neftgeophysika, Moormansk Neftgeophysika, Central

Geophysical Expedition, VNIIGAS, Sever Gazprom, Orenburg Gazprom, Lukoil Kogalym, Yukos Moscow, Luk-Arco, Slavneft Tver, etc., to name a few, are the leading computing centers installed by ROY International in the oil and gas field. At present, ROY International is completing the final installation at Tatneftgeophysika, one of the major geophysical companies of Russia. Within the framework of this project, ROY International is finalizing the execution of the installation of a Sun Supercomputer together with leading software in the world. This complex is specially designed for 3D geophysical data processing. The world's leading oil and gas producers love the characteristics of Enterprise10000, also called the 'CRAY-Killer'. Starting with new 18 Ultra SPARC-II Microprocessors, TByte Storage Array, TByte ETL Tape Library, more than 100 high power Sun Work Stations and Networking etc., this center is the most powerful computing center in Russia and all CIS countries. Being the Systems Integrator, ROY International, after several negotiations, selected Paradigm Geophysical software (Israel) for Data Processing and Schlumberger GeoQuest Software (France) for Data Interpretation. This is in addition to the existing Data Processing Software, Landmark, of America. ROY International also has agreements with manufacturers like Storage Tech (USA), Qualstar Corporation (USA), E.R.Mapping (Australia), M4 Data Ltd., (England), Fujitsu (Japan), 3Com (USA), OYO Plotters and Instruments (Japan), etc., and also with various Russian manufacturers and software developers. Trained Specialists and Engineers of ROY International completed the networking job within two weeks and software installation and training just completed. Four of the ROY International's specialists are with Ph.D. qualifications in this field. Processing and Interpretation of Data are handled

by more than 400 highly qualified employees of Tatneftgeophysica. The General Director of Tatneftgeophysika, Mr.Rinat Kharisov said, "This is the second time we are entering into a new era of major technological modernization of their computing center, which is being executed by ROY International. Six years back, we have modernized our computing center with the help of ROY International. They replaced the Russian ES computers with the Sun SPARC Center2000 which increased our computing power to 20 times. The present installation increases our power to another 70 times. This enables us to find the results of Interpretation Data, saving substantial time and money and we can compete at the global market". "The new Super Computer project once more confirms that the trend in Russia is to set up large scale information centers, for which high end Super Computers are required", said Dr. Cherian Eapen, President of ROY International. "It was a difficult task to get licenses from all US Govt. Departments. The application for export license and non-proliferation compliance letter etc. were routed through the Russian Ministry of Fuel and Energy and through the US Embassy in Moscow to the Bureau of Exports Administration in Washington DC. The procedure took a long time to grant permission to allow the use of the Supercomputer for a civilian customer in Russia as Russia is still under the list of countries for nuclear proliferation concerns. Since ROY International has got a clean record and doesn't have any military deals and since it is strict in working only for a peaceful production activities, it got an advantage on license application. Departments of Commerce, State, Defense, Atom Energy Ecology cleared the license application for this Super Computer earlier and finally, the Department of Energy also gave the green light to lift it to Russia. This 2 Tons Super Computer complex was lifted from San Francisco to Amsterdam by Lufthansa Cargo and from there to Moscow and to Nabereshni

Chelni, a nearby airport of the end user in Tatarstan Republic by a chartered flight arranged by ROY International. Dr.Cherian Eapen said, "With the highest of security safeguard procedures, we were able to reach the System to the pre-approved and designed site of the computing center, per license requirement. Every moment was tense due to danger of security reasons against any physical diversion during shipment. One of our employees from Moscow, Victor, traveled with the freight forwarders and security crew and informed me each hour the progress of loading and off loading and air and ground transport to the destination. About 4 AM on December 25, Christmas day morning, I got the final call for that day from Victor, asking me to take rest as the job is completed and therefore requesting to allow them to celebrate the installation of the Super Computer, by opening a bottle of Vodka."

BiO News

Posted By: gmax Date: 10/06/00 16:48 Summary:BioInform: IBM Designs Architecture for Blue Gene Supercomputer, Collaborates with Academia ``Since IBM's announcement last year that it would spend $100 million to build a supercomputer called Blue Gene for protein folding research, it has begun collaborating with scientists at Indiana University, Columbia University, and the University of Pennsylvania on some of the mathematical techniques and software needed for the system. ``The company has also decided to use a cellular architecture for the machine, where it will use simple pieces and replicate them on a large scale. Protein folding research requires advances in computational power and molecular dynamics techniques - the mathematical method for calculating the movement of atoms in the formation of proteins, said Joe Jasinski, IBM's newly appointed senior manager of Blue Gene and the computational biology center. ```The first problem that we are attacking with Blue Gene is to understand at the atomic level the detailed dynamics, the motions involved in protein folding,' Jasinski said. `That's a very computationally intensive problem which requires at least a petaflop computer and probably something bigger.' ``Most of the system software as well as the routines that will drive the applications are being developed by IBM's computational biology group, which was formed in 1992 and now numbers about 40 scientists and engineers.''

DIFFERENT SUPER COMPUTER

Canada's Fastest Computer Simulates Galaxies In Collision
by Nicolle Wahl Toronto - Jul 25, 2003

A $900,000 supercomputer at the University of Toronto -- the fastest computer in Canada -- is heating up astrophysics research in this country and burning its way up the list of the world's fastest computers. The new computer, part of the Department of Astronomy and Astrophysics and the Canadian Institute for Theoretical Astrophysics (CITA), was ranked as the fastest computer in Canada and the
Shortened sequence of images showing the detailed interaction of two galaxies colliding

39th fastest in the world in the latest list from www.top500.org, compiled by the Universities of Mannheim and Tennessee and the National Energy Research Scientific Computing Center at Lawrence Berkeley National Laboratory. "An essential element of modern astrophysics is the ability to carry out largescale simulations of the cosmos, to complement the amazing observations being undertaken," said Professor Peter Martin, chair of astronomy and astrophysics and a CITA investigator. "With the simulations possible on this computer, we have in effect a laboratory where we can test our understanding of astronomical phenomena ranging from the development of structure in the universe over 14 billion years to the development of new planets in star-forming systems today." When the computer, created by the HPC division of Mynix Technology of Montreal (now a part of Ciara Technologies), starts its calculations, the 512 individual central processing units can heat up to 65 C, requiring extra ventilation and air-conditioning to keep the unit functioning. But with that heat comes the capability of performing more than one trillion calculations per second, opening the door to more complex and comprehensive

simulations of the universe. It is the only Canadian machine to break the Teraflop barrier -- one trillion calculations per second -- and it's the fastest computer in the world devoted to a wide spectrum of astrophysics research. "This new computer lets us solve a variety of problems with better resolution than can be achieved with any other supercomputer in Canada," said Chris Loken, CITA's computing facility manager. "Astrophysics is a science that needs a lot of computer horsepower and memory and that's what this machine can provide. The simulations are also enabled by in-house development of sophisticated parallel numerical codes that fully exploit the computer's capabilities." The machine, nicknamed McKenzie (after the McKenzie Brothers comedy sketch on SCTV), with 268 gigabytes of memory and 40 terabytes of disk space, consists of two master nodes (Bob and Doug), 256 compute nodes, and eight development nodes. All of these are networked together using a novel gigabit networking scheme that was developed and implemented at CITA. Essentially, the two gigabit Ethernet ports on each node are used to create a "mesh" that connects every machine directly to another machine and to one of 17 inexpensive gigabit switches. It took four people about two days and two kilometres of cable to connect this network. The unique CITA design drives down the networking cost in the computer by at least a factor of five and the innovative system has attracted industry attention. Professor John Dubinski has used the new computer to examine both the formation of cosmological structure and the collisions of galaxies by simulating the gravitational interaction of hundreds of millions of particles representing stars and the mysterious dark matter. The anticipated collision of the Milky Way Galaxy with our neighbouring Andromeda galaxy -- an event predicted to take place in three billion years time -- has been modeled at unprecedented resolution.

New simulations on the formation of supermassive black holes, again with the highest resolution to date, have been carried out by his colleagues Professors Ue-Li Pen and Chris Matzner. They have already uncovered clues which may explain the mystery of why the black hole at the center of our galaxy is so much fainter than had been expected theoretically. The team has even grander plans for the future. "In astrophysics at the University of Toronto we have continually exploited the latest computing technology to meet our requirements, always within a modest budget," Martin said. "This is a highly competitive science and to maintain our lead we are planning a computer some ten times more powerful."

China To Build World's Most Powerful Computer
Beijing (Xinhua) Jul 29, 2003

The Downing Information Industry Co., Ltd., a major Chinese manufacturer of highperformance computers, is to build the world's most powerful computer, capable of performing 10 trillion calculations per second. Scheduled to be completed by March next year, the super computer marks China's first step in the development of a cluster computer system, which has the highest calculation speed in the world, according to a source with the company. Previously, the Downing Information Industry Co., Ltd. had successfully developed a super computer, capable of performing 4 trillion calculations per second.
China's future science and defence needs will require ever more powerful high performance computer systems.

Code-named "Shuguang4000A", the planned super computer covers an area equal to a quarter of a football field, and it will use processors developed by AMD, a United States computer chip maker. AMD and the Chinese company have signed a cooperation agreementto develop the planned super computer of China. The fastest existing cluster computer system in the world is capable of calculating at a speed of 7.6 trillion bytes per second.

First Super Computer Developed in China
China's first super computer which is capable of making 1.027 trillion calculations per second showed up in Zhongguancun, known as a "Silicon Valley" in the Chinese capital Beijing, Thursday. The computer, developed by the Legend Group Corp., China's leading computer manufacturer, boasts the same operation speed as the 24th fastest computer in the world's top 500 super computers. The leading 23 super computers were developed by Japan and the United States respectively. Legend president Yang Yuanqing said the computer will be installed at the mathematics and system science research institute affiliated to the Chinese Academy of Sciences in early September. It will be used in calculating hydromechanics, disposing petroleum and earth quake materials, climatic mode calculation, materials science calculation, DNA and protein calculation. Yang said it only takes the computer two minutes to complete the simulation of global climatic changes in one day, compared with 20 hours by other large computers. Computers with super calculation speed used to be the tool of small number of scientists in labs in the past, but now they are widely used in economic and social fields, even in film-

making. A computer, which is capable of making 85.1 trillion calculations per second, the highest calculation speed in the world, has been recently developed in Japan.

Shell to use Linux supercomputer for oil quest
December 12, 2000 Web posted at: 9:00 AM EST (1400 GMT)

LONDON, England (Reuters) -- Linux, the free computer operating system, is expected to win another high-profile victory on Tuesday when Anglo-Dutch oil company Royal Dutch/Shell will announce it is going to install the world's largest Linux supercomputer. Shell's Exploration & Production unit will use the supercomputer, consisting of 1,024 IBM X-Series servers, to run seismic and other geophysical applications in its search for more oil and gas.

Data collected in Shell exploration surveys will be fed into the computer, which will then analyze it. The announcement comes just days after Swedish telecom operator Telia said it would use a large Linux mainframe to serve all its Internet subscribers, replacing a collection of Sun Microsystems servers. "Linux is coming of age," said one source close to the deal. Linux, developed by the Fin Linus Torvalds and a group of volunteers on the Web, has been embraced by International Business Machines Corp. as a flexible alternative to licensed software systems such as Microsoft's Windows or the Unix platforms. With Linux companies can quickly add or remove computers without worrying about licenses for the operating software. Over the past year the software has been tested and trialled for business critical applications. Major deals have now started to come through. Recently Musicland Stores Corp., the U.S. company that owns Sam Goody, said it would install new Linux and Java-based cash registers. The most recent announcements indicate that Linux usage is becoming more versatile, with the operating system moving into many different applications, not just Internet computers.

World's fastest computer simulates Earth
Saturday, November 16, 2002 Posted: 2:57 PM EST (1957 GMT)

SAN JOSE, California (AP) -- A Japanese supercomputer that studies the climate and other aspects of the Earth maintained its ranking as the world's fastest computer, according to a study released Friday.
The Earth Simulator in Yokohama, Japan, performs 35.86 trillion calculations per second -- more than 4 1/2 times greater than the next-fastest machine. Earth Simulator, built by NEC and run by the Japanese government, first appeared on the list in June. It was the first time a supercomputer outside the United States topped the list. Two new machines, called "ASCI Q," debuted in the No. 2 and No. 3 spots. The computers, which each can run 7.73 trillion calculations per second, were built by Hewlett-Packard Co. for Los Alamos National Laboratory in New Mexico.

The Earth Simulator consists of 640 supercomputers that are connected by a high-speed network.

Clusters of personal computers rank
Story Tools For the first time, high-performance machines built by clustering personal computers appeared in the top 10. A system built by Linux NetworX and Quadrics for Lawrence Livermore National Laboratory ranked No. 5. A system built by High Performance Technologies Inc. for the National Oceanic and Atmospheric Administration's Forecast Systems Laboratory was No. 8. Hewlett-Packard Co. led with 137 systems on the list, followed by RELATED International Business Machines Corp. with 131 systems. No. 3 Sun Top 500 most powerful computers Microsystems Inc. built 88 of the top 500 systems. The Top 500 list, which has been released twice annually since 1993, is compiled by researchers at University of Mannheim, Germany; the Department of Energy's National Energy Research Scientific Computing Center in Berkeley and the University of Tennessee.

G5 SUPERCOMPUTER IN THE WORKS
Interesting rumor. Virginia Tech got bumped to the head of the line with their order of 1100 new G5 computers, so that they could build a supercomputer and make Linpack's Top 500 list this year.

Not too surprising that Apple gave them preferential treatment. Wonder if Apple might be tempted into making a commercial? Even if they just posted it on-line, it might prove interesting.

Virginia Tech building supercomputer G5 cluster
By Nick dePlume , Publisher and Editor in Chief August 30, 2003 - Virginia Tech University is building a Power Mac G5 cluster that will result in a supercomputer estimated to be one of the top five fastest in the world. In yesterday's notes article , we reported that Virginia Tech had placed a large order of dual-2GHz G5s to form a cluster. Since that time, we've received additional information, allowing us to confirm a number of details. According to reports, Virginia Tech placed the dual-2GHz G5 order shortly after the G5 was announced. Multiple sources said Virginia Tech has ordered 1100 units; RAM on each is said to be upgraded to 4GB or 8GB. The G5s will be clustered using Infiniband to form a 1100-node supercomputer delivering over 10 Teraflops of performance. Two sources said the cluster is estimated to be one of the top five fastest supercomputers in the world. However, Virginia Tech's on a deadline. The university needs to have the cluster completely set up this fall so that it can be ranked in Linpack's Top 500 Supercomputer list. Apple bumped Virginia Tech's order to the front of the line -- even in front of first day orders -- to get them out the door all at once. Sources originally estimated the G5s will arrive the last week of August; they're still on track to arrive early, possibly next week. This information is more-or-less public within the university community but no announcement has been made. Earlier in the month, Think Secret contacted Virginia Tech's Associate Vice President for University Relations, who said the report was an "interesting story" and agreed to see what he could confirm. The university didn't respond to follow-up requests for comment.

INDIA's 'PARAM-10,000' SUPER COMPUTER
INDIA'S AMAZING PROGRESS IN THE AREA OF HI-TECH COMPUTING"

India's Hi-Tech expertise, has made Harare Pyare Bharat, a nation to reckon with. Following, is a short article by Radhakrishna Rao, a freelance writer who has contributed this material to "INDIA - Perspectives" (August 1998), page 20 :"The restrictions imposed by the United States of America on the transfer of knowhow in frontier areas of Technology, and its consistent refusal to make available to India a range of hardware for its development, have proved to be a blessing in disguise, because Indian scientists and engineers have now managed to develop, indigenously, most of the components and hardware required for its rapidly advancing space and nuclear power programmes. It was again the refusal of the U.S. administration to clear the shipment to India of a Cray X-MP super computer, for use by the Institute of Sciences (IISc.), Bangalore, in the 1980's, along with severe restrictions on the sale of computers exceeding 2000 Mega Theoretical Operations per Second (MTOPS), that led India to build one of the most powerful super computers in the world. In fact, the unveiling of the "PARAM-10,000" super-computer, capable of performing one trillion mathematical calculations per second, stands out as a shining example of how 'restrictions and denials' could be turned into impressive scientific gains. For the Pune-based Centre for Development of Advanced Computing (C-DAC), which built this super-computing machine, it was a dream come true. In fact, the "PARAM-10,000", based on an open-frame architecture, is considered to be the most powerful super-computer in Asia, outside Japan. So far, only U.S.A. and Japan have built up a proven capability to build similar types of supercomputers. To be sure, Europe is yet to build its own super-computer in this category. As it is, "PARAM-10,000", has catapulted India into the ranks of the elite nations that, already, are in the rarefied world of tera flop computing which implies a capability to perform one trillion calculations per second. In this context, a beaming Dr. Vijay P. Bhatkar, Director, of C-DAC, says, "We can now pursue our own mission critical problems at our own pace and on our own terms. By developing this, India's esteem in Information Technology (IT) has been further raised."

As things stand now, "PARAM-10,000" will have applications in as diverse areas as long-range weather forecasting, drug design, molecular modelling, remote sensing and medical treatment. According to cyber scientists, many of the complex problems that India's space and nuclear power programmes may encounter in the future could be solved with "PARAM-10,000", without going in for the actual ground level physical testing. On a more practical plain, it could help in the exploration of oil and gas deposits in various parts of the country. Perhaps the post exciting application of "PARAM-10,000" will be in storing information on Indian culture and heritage, beginning with Vedic times. "We want to preserve our timeless heritage in the form of a multimedia digital library on "PARAM-10,000", says Dr. Bhatkar, That C-DAC could manage to build a "PARAM-10,000" machine in less than five years is a splendid tribute to the calibre and dedication of its scientists and engineers. No wonder C-DAC has bagged orders for as many as three "PARAM-10,000" machines. And two of these are from abroad ; a Russian academic institute and Singapore University are keenly awaiting the installation of "PARAM-10,000" machines on their premises. The third machine will be used by the New Delhi-based National Informatics Centre (NIC) for setting up a geomatics faculty designed to provide solutions in the area of remote sensing and image processing. C-DAC is also planning to develop advanced technologies for the creation of a national information infrastructure. Meanwhile, C-DAC has proposed the setting up of a full fledged company to commercially exploit the technologies developed by it. C-DAC was set up in 1988 with the mandate to build India's own range of super-computers. Incidently, "PARAM-10,000" is a hundred times more powerful that the first Param machine built, way back in the early 1990's." --- (Radhakrishna Rao , Author)

IBM plans world's most powerful Linux supercomputer
IDG News Service 7/30/03

A Japanese national research laboratory has placed an order with IBM Corp. for a supercomputer cluster that, when completed, is expected to be the most powerful Linux-based computer in the world. The order, from Japan's National Institute for Advanced Industrial Science and Technology (AIST), was announced by the company on Wednesday as it simultaneously launched the eServer 325 system on which the cluster will be largely based. The eServer 325 is a 1U rack mount system that includes two Advanced Micro Devices Inc. Opteron processors of either model 240, 242 or 246, said IBM in a statement. The supercomputer ordered by AIST will be built around 1,058 of these eServer 325 systems, to make a total of 2,116 Opteron 246 processors, and an additional number of Intel Corp. servers that include a total of 520 of the company's third-generation Itanium 2 processor, also known by its code name Madison. The Opteron systems will collectively deliver a theoretical peak performance of 8.5 trillion calculations per second while the Itanium 2 systems will add 2.7 trillion calculations per second to that for a total theoretical peak performance for the entire cluster of 11.2 trillion calculations per second. That would rank it just above the current most powerful Linux supercomputer, a cluster based on Intel's Xeon processor and run by Lawrence Livermore National Laboratory (LLNL) in the U.S. That machine has a theoretical peak performance of 11.1 trillion calculations per second, according to the latest version of the Top 500 supercomputer ranking. Based on that ranking, the new machine would mean Japan is home to two out of the three most powerful computers in the world. The current most powerful machine, the NEC Corp.-built Earth Simulator of the Japan Marine Science and Technology Center, has a theoretical peak performance of 41.0 trillion calculations per second while that of the second-fastest machine, Los Alamos National Laboratory's ASCI Q, is 20.5 trillion calculations per second. The eServer 325 can run either the Linux or Windows operating systems and the supercomputer ordered by AIST will run SuSE Linux Enterprise Server 8. IBM said it expects to deliver the cluster to AIST in March, 2004. AIST will link the machine with others as part of a supercomputer grid that will be used in research of grid technology, life sciences bioinformatics and nanotechnology, IBM said. General availability of the eServer 325 is expected in October this year and IBM said prices for the computer start at US$2,919. The computers can also be accessed through IBM's on-demand service where users pay for processing power based on capacity and duration.

IBM's announcement is the second piece of good news for AMD and its Opteron processor within the last two weeks. The processor, which can handle both 32-bit and 64-bit applications, was launched in April this year. China's Dawning Information Industry Co. Ltd. announced plans last week to build a supercomputer based on AMD's Opteron processor. The Dawning 4000A will include more than 2,000 Opteron processors, with a total of 2T bytes of RAM and 30T bytes of hard-disk space and is expected to deliver performance of around 10 trillion calculations per second. The Beijing-based company has an order for the machine but has not disclosed the name of the buyer or when the computer will be put into service. Opteron processors were also chosen for a supercomputer which is likely to displace the AIST machine as the most powerful Linux supercomputer. Cray Inc. is currently constructing a Linux-based supercomputer called Red Storm that is expected to deliver a peak performance of 40 trillion calculations per second when it is delivered in late 2004. Linux developer SuSE is also working with Cray on that machine.

Jefferson Team Building COTS Supercomputer
Newport News - Jul 03, 2003

Scientists and engineers from Jefferson Lab’s Chief Information Office have created a 'cluster supercomputer' that, at peak operation, can process 250 billion calculations per second Science may be catching up with video gaming. Physicists are hoping to adapt some of the most potent computer components developed by companies to capitalize on growing consumer demands for realistic simulations that play out across personal computer screens.
Chip Watson, head of the HighPerformance Computing Group (from left); watches Walt Akers, computer engineer: and Jie Chen, computer scientist, install a Myrinet card into a computer node.

For researchers, that means more power, less cost, and much faster and more accurate calculations of some of Nature's most basic, if complex, processes. Jefferson Lab is entering the second phase of a three-year effort to create an offthe-shelf supercomputer using the next generation of relatively inexpensive, easily available microprocessors. Thus far, scientists and engineers from JLab's Chief Information Office have created a "cluster supercomputer" that, at peak operation, can process 250 billion calculations per second. Such a 250 "gigaflops" machine -- the term marries the nickname for billion to the abbreviation for "floating-point operations" -- will be scaled up to 800 gigaflops by June, just shy of one trillion operations, or one teraflop. The world's fastest computer, the Earth Simulator in Japan, currently runs at roughly 35 teraflops; the next four most powerful machines, all in the United States, operate in the 5.6 to 7.7 teraflops range.

The Lab cluster-supercomputer effort is part of a broader collaboration between JLab, Brookhaven and Fermi National Laboratories and their university partners, in a venture known as the Scientific Discovery through Advanced Computing project, or SciDAC, administered by the Department of Energy's Office of Science. SciDAC's aim is to routinely make available to scientists terascale computational capability. Such powerful machines are essential to "lattice quantum chromodynamics," or LQCD, a theory that requires physicists to conduct rigorous calculations related to the description of the strong-force interactions in the atomic nucleus between quarks, the particles that many scientists believe are one of the basic building blocks of all matter. "The big computational initiative at JLab will be the culmination of the lattice work we're doing now," says Chip Watson, head of the Lab's High-Performance Computer Group. "We're prototyping these off-the-shelf computer nodes so we can build a supercomputer. That's setting the stage for both hardware and software. " The Lab is also participating in the Particle Physics Data Grid, an application that will run on a high-speed, high-capacity telecommunications network to be deployed within the next three years that is 1,000 times faster than current systems. Planners intend that the Grid will give researchers across the globe instant access to large amounts of data routinely shared among far-flung groups of scientific collaborators. Computational grids integrate networking, communication, computation and information to provide a virtual platform for computation and data management in the same way that the Internet permits users to access a wide variety of information.

Whether users access the Grid to use one resource such as a single computer or data archive, or to use several resources in aggregate as a coordinated, virtual computer, in theory all Grid users will be able to "see" and make use of data in predictable ways. To that end, software engineers are in the process of developing a common set of computational, programmatic and telecommunications standards. "Data grid technology will tie together major data centers and make them accessible to the scientific community," Watson says. "That's why we're optimizing cluster-supercomputer design: a lot of computational clockspeed, a lot of memory bandwidth and very fast communications." Computational nodes are key to the success of the Lab's cluster supercomputer approach: stripped-down versions of the circuit boards found in home computers. The boards are placed in slim metal boxes, stacked together and interconnected to form a cluster. Currently the Lab is operating a 128-node cluster, and is in the process of procuring a 256-node cluster. As the project develops, new clusters will be added each year, and in 2005 a single cluster may have as many as 1,024 nodes. The Lab's goal is to get to several teraflops by 2005, and reach 100 teraflops by 2010 if additional funding is available. "[Our cluster supercomputer] is architecturally different from machines built today," Watson says. "We're wiring all the computer nodes together, to get the equivalent of three-dimensional computing." That can happen because of continuing increases in microprocessor power and decreases in cost. The Lab's approach, Watson explains, is to upgrade continuously at the lowest cost feasible, replacing the oldest third of the system each year.

Already, he points out, the Lab's prototype supercomputer is five times cheaper than a comparable stand-alone machine, and by next year it will be 10 times less expensive. Each year as developers innovate, creating more efficient methods of interconnecting the clusters and creating better software to run LQCD calculations, the Lab will have at its disposal a less expensive but more capable supercomputer. "We're always hungry for more power and speed. The calculations need it," Watson says. "We will grow and move on. The physics doesn't stop until we get to 100 petaflops [100,000 teraflops], maybe by 2020. That's up to one million times greater than our capability today. Then we can calculate reality at a fine enough resolution to extract from theory everything we think it could tell us. After that, who knows what comes next?"

SUN Genome Center SuperComputing
The Genome Center of Wisconsin in May 2000 opened a new supercomputer facility built around a Sun Microsystems Enterprise 10000 and several smaller computers. The E 10000 is designed to provide flexible computational power using from between 1 and 36 processors as needed configurable on the fly. In addition to its processor power it has 36 gigabytes of memory and 3 Terabytes of disk storage to provide the optimal computing environment for genomic research. In future the E 10000 will be able to expand to 64 processors, 64 gigabytes of ram and 60 terabytes of online disk storage. On September 22, 2000 Sun Microsystems announced the Genome Center SuperComputing was being named a Sun Center of Excellence. Being a Center of Excellence is a statement that Sun Acknowledges we are a quality center of computing and that there is a continuing partnership between the Genome Center and Sun Microsystems.

Mission

The mission of Genome Center SuperComputing is to provide Genomic researchers and their academic collaborators access to computing power that would otherwise be outside the scope of their organizations. In providing access to computing power, storage, local databases and most of the commonly available Unix based biological software we are trying to keep researchers from working with inadequate resources or supporting unwanted infrastructure.

Cray super Computer

Applications for Cray Systems
The Cray Applications Group is committed to making available the software which is important to our customers. Cray Inc. works with third-party software vendors to port codes and to assure that our customers get the best possible performance. From bioinformatics to seismic imaging to automotive crash simulations, Cray systems are used to run applications which solve both large and complex computational problems. Cray applications data sheets: AMBER and Cray Inc. (pdf) Gaussian 98 and Cray Inc. Supercomputers (pdf) MSC.Nastran and Cray Inc. Supercomputers (pdf) MSC.Nastran Performance Enhancements on Cray SV1 Supercomputers (pdf)

Cray Professional Services

For more than 25 years, Cray has been at the forefront of high performance computing (HPC), contributing to the advancement of science, national security, and the quality of human life. Cray has designed, built, and supported high-performance computing solutions for customers all around the world. Cray helps ensure the success of supercomputer implementation by partnering with customers to provide complete solutions for the most challenging scientific and engineering computational problems. These robust solutions utilize Cray's deep supercomputing expertise and sterling reputation for quality. Cray's understanding of high-performance computing is unrivaled. Our Professional Services Solutions give you access to some of the most savvy, experienced minds in computing. Examples of capabilities in this area include software development, custom hardware, extensions to Cray supercomputing products, and access to the systems in Cray's world-class data center. We help Cray customers in all aspects of high-performance computing, from problem analysis to solution implementation. Cray Professional Services draws on Cray's extensive talent and expertise company-wide. Why engage Cray Professional Services?  Over 25 years of experience in the HPC industry World-class technical expertise with access to the best minds, methods, and tools in the industry  Exceptional customer service and dedication to quality



STORAGE SERVICES Cray Professional Services provides SNIA certified SAN specialists to deliver solutions related to high performance data storage, including SAN design and implementation. Storage services include StorNext File System and StorNext Management Suite implementations, RS200 extensions, custom Cray SANs, and legacy data migrations. CUSTOM ENGINEERING Cray has gathered some of the best engineering minds and technologies in the world to produce its computer systems. To achieve the extreme levels of performance found in supercomputers requires an enormous breadth and depth of leading-edge technical talent. This talent is transferable into other high-performance applications as well in terms of system design, code porting and optimization, system packaging, system power and cooling technologies, and troubleshooting issues in the design and manufacturing process. Cray Custom Engineering also offers custom design enhancements to existing Cray products and the use of traditional Cray hardware as embedded components in a variety of other applications and products. The custom engineering offering from Cray is targeted to assist both traditional and nontraditional Cray customers in addressing their most extreme technical issues.

CRAY CONSULTING Cray customers address the most complex, highperformance computing problems. Whether in support of issues of national security, safety, design simulation, or the environment, Cray systems have been the favored computational solution for more than 25 years. To produce these state of the art systems, Cray has developed a broad spectrum of core competencies in the design, implementation, and optimization of highperformance computing solutions. Cray scientists and engineers are 100% focused on high-performance problems and solutions - this is our business. Cray now offers this tremendous intellectual capital to our customers to address your needs. SUPERCOMPUTING ON DEMAND Several generations of Cray products are available to support your highperformance computing needs. "On demand" means these resources can be scheduled for use whenever and wherever you need them. Whether it's providing compute services to cover a peak in operational demand, support for application development or code optimization, or an ASP-based environment, Cray will work with you to make computational resources available to meet your specific high-performance computing needs. CRAY TRAINING Cray products are designed to fit the highest performance compute needs of our customers. Our goal is to ensure that our customers make the most of

their systems. Our training options are designed to enable Cray customers to see a quick return on their compute investment. Classes are available on a wide variety of topics and platforms, such as system administration, programming and optimization, and various quick-start packages. SITE ENGINEERING Cray has been installing, relocating, and optimizing computing environments for over 25 years. Managing on-site system power and cooling, and interior climate conditions requires the skills of highly trained personnel to ensure optimal system support and performance. Site Engineering at Cray merges the needs and dimensions of a customer's specific computing environment and translates them into comprehensive work plans and complete site engineering solutions.

Software for Cray Systems
Powerful hardware systems alone cannot meet the requirements of the most demanding scientific and engineering organizations. Equally powerful, robust software is needed to turn supercomputers into indispensable productivity tools for the sophisticated government, commercial, and academic user communities. In these demanding environments, where multimillion-dollar projects are at stake, reliability, resource management, single job performance, complex multijob throughput, and highbandwidth data management are critical.  UNICOS® The undisputed leader among high-end supercomputer operating systems  UNICOS/mk™ The UNICOS/mk operating system fully supports the Cray T3E system's globally scalable architecture  CF90® Programming Environment The CF90 Programming Environment consists of an optimizing Fortran compiler, libraries, and tools  Cray C++ Programming Environment C++ and C are the computer languages used today for many highperformance applications  Message-Passing Toolkit (MPT) Provides optimized versions of industry-standard message-passing libraries and software  Network Queuing Environment (NQE) Workload management environment that provides batch scheduling and interactive load balancing



Distributed Computing Environment (DCE) Distributed File Service (DFS) An industry-standard, vendor-neutral set of tools and services providing distributed computing capability. DFS is a distributed DCE application providing an integrated file system with a unified name space, secure access, and file protection.  Data Migration Facility (DMF) A low-overhead hierarchical storage management (HSM) solution  Cray/REELlibrarian A volume management system that controls libraries of tape volumes

Cray Systems at Work
Cray systems provide powerful high performance solutions for the world's most complex computational problems. The sustained performance obtained from Cray supercomputers is used by researchers and computer scientists spanning such varied disciplines as automotive manufacturing, geological sciences, climate prediction, pharmaceutical development, and national security. Cray supercomputers are used worldwide in research, academia, industry, and government.

The Road to La-La Land - Pittsburgh Supercomputing Center researcher Pei Tang uses the Cray T3E to probe the mysteries of anesthesia. Biomedical Modeling at the National Cancer Institute Researchers from around the world use NCI's Cray SV1 system to solve some of the most difficult problems in computational biology -- studying protein structure and function at the most detailed levels.

Clean Power - George Richards, leader of the National Energy Technology Laboratory's combustion dynamics team, takes on the challenge of converting fuel to energy without creating pollutants by using simulations on PSC's Cray T3E. A Thumb-Lock on AIDS - PSC's Marcela Madrid simulates an HIV enzyme on the Cray T3E to help develop drugs that shut down HIV replication.

SUPER COMPUTERS
There are two main kinds of supercomputers: vector machines and parallel machines. Both kinds work FAST, but in different ways. Let's say you have 100 math problems. If you were a vector

computer, you would sit down and do all the problems as fast as you could. To work like a parallel computer, you would get some friends and share the work. With 10 of you, you would each do 10 problems. If you got 20 people, you'd only have to do 5 problems each. No matter how good you are at math, it would take you longer to do all 100 problems than to have 20 people do them together.

CRAY T3E
SDSC's newest supercomputer is the CRAY T3E. The T3E is a parallel supercomputer and has 256 processors to work on problems. (Let's not worry about what "T3E" stands for.) If you get all the T3E processors going full speed, it can do 153.4 billion -- that's 153,400,000,000 -- math calculations every second. But researchers usually only use some of the T3E's processors at once. That way, many researchers can run their programs at the same time.

CRAY C90: Vector Machine
The CRAY C90 is the busiest of SDSC's supercomputers and cost $26 million. A problem that takes a home computer 8 hours to solve, the CRAY C90 can do in 0.002 seconds. And some scientists have problems that take the CRAY C90 a couple DAYS to do. The CRAY C90 is a vector machine with eight processors -- eight vector machines in one. With all eight processors, the CRAY C90 can do 7.8 gigaFLOPS. (In peoplepower you'd need one and a half times the

Earth's population.) A pretty slick Pentium PC might reach about 0.03 gigaFLOPS, depending on who you ask.

Application Of Super Computer]
Decision Agent Offers Secure Messaging At Pentagon
Woodland Hills - May 07, 2003

Northrop Grumman Corporation went live at the Pentagon April 1 with the secure organizational messaging services of the "Decision Agent" following several months of installation and testing by the company's California Microwave Systems business unit. Installed and managed by the Pentagon no Max it's a smoking zone, Telecommunications Center (PTC), which supports not the secure communication bubble over 30,000 users and 1,000 organizations, "Decision Agent" is designed to provide an enterprise-wide information profiling and management portal for the Defense Messaging System (DMS) Version 3.0. "The 'Decision Agent' has been instrumental in allowing us to provide protected message traffic for our customers," said Marvin Owens, director of the PTC. "It has proved to be a tremendous tool in helping us achieve our mission of implementing DMS in the Pentagon as well as for our other customers worldwide." The new system at the PTC supports DMS communications for the Office of the Secretary of Defense, the Army Operations Center and the military services headquarters' staffs. In addition, it provides a communication gateway that permits interoperability between Department of Defense organizations that use DMS and allied NATO and non-Defense Department organizations that use legacy systems. "By enabling DMS messaging without FORTEZZA cards and readers, and by eliminating the need to install add-ons to end-user devices, 'Decision Agent' will allow the Pentagon and other government agencies to reduce the costs and manpower requirements traditionally associated with DMS implementations," said

John Haluski, vice president of California Microwave Systems' Information Systems unit. The "Decision Agent" consists of a suite of integrated software applications that run on a Windows 2000 server. These include Northrop Grumman's LMDS MailRoom, the most powerful profiling engine currently available, and Engenium Corporation's Semetric, a knowledge-based retrospective search engine. The system enhances DMS functionality by automating the processes of identifying, filtering and distributing military organization messages to specified addresses and recipients, based on interest profiles and security clearances. "Decision Agent" provides other enhancements as well, including virus checks, security checks for possible mislabeling of messages and attachments, Web-based message preparation and Boolean logic keyword and concept searches.

Frozen Light May Make Computer Tick Later This Century
Boston - May 22, 2003

NASA-funded research at Harvard University, Cambridge, Mass., that literally stops light in its tracks, may someday lead to breakneck-speed computers that shelter enormous amounts of data from hackers. The research, conducted by a team led by Dr. Lene Hau, a Harvard physics professor, is one of 12 research projects featured in a special edition of Scientific American entitled "The Edge of Physics," available through May 31. In their laboratory, Hau and her colleagues have been able to slow a pulse of light, and even stop it, for several- thousandths of a second. They've also created a roadblock for light, where they can shorten a light pulse by factors of a billion. "This could open up a whole new way to use light, doing things we could only imagine before," Hau said. "Until now, many technologies have been limited by the speed at which light travels."
an old thumper at NASA in the early 1960s

The speed of light is approximately 186,000 miles per second (670 million miles per hour). Some substances, like water and diamonds, can slow light to a limited extent. More drastic techniques are needed to dramatically reduce the speed of light. Hau's team accomplished "light magic" by laser-cooling a cigar-shaped cloud of sodium atoms to one- billionth of a degree above absolute zero, the point where scientists believe no further cooling can occur. Using a powerful electromagnet, the researchers suspended the cloud in an ultra-high vacuum chamber, until it formed a frigid, swamp-like goop of atoms. When they shot a light pulse into the cloud, it bogged down, slowed dramatically, eventually stopped, and turned off. The scientists later revived the light pulse and restored its normal speed by shooting an additional laser beam into the cloud. Hau's cold-atom research began in the mid-1990s, when she put ultra-cold atoms in such cramped quarters they formed a type of matter called a BoseEinstein condensate. In this state, atoms behave oddly, and traditional laws of physics do not apply. Instead of bouncing off each other like bumper cars, the atoms join together and function as one entity. The first slow-light breakthrough for Hau and her colleagues came in March 1998. Later that summer, they successfully slowed a light beam to 38 miles per hour, the speed of suburban traffic. That's two million times slower than the speed of light in free space. By tinkering with the system, Hau and her team made light stop completely in the summer of 2000. These breakthroughs may eventually be used in advanced opticalcommunication applications. "Light can carry enormous amounts of information through changes in its frequency, phase, intensity or other properties," Hau said.

When the light pulse stops, its information is suspended and stored, just as information is stored in the memory of a computer. Light-carrying quantum bits could carry significantly more information than current computer bits. Quantum computers could also be more secure by encrypting information in elaborate codes that could be broken only by using a laser and complex decoding formulas. Hau's team is also using slow light as a completely new probe of the very odd properties of Bose-Einstein condensates. For example, with the light roadblock the team created, they can study waves and dramatic rotating-vortex patterns in the condensates.

Navy to use IBM supercomputer for storm forecasting
By Reuters August 2, 2000, 5:40 PM PT http://news.com.com/2100-1001-244009.html?tag=prntfr NEW YORK--IBM said today the U.S. Department of Defense paid $18 million for one of the world's fastest supercomputers to help Navy vessels avoid maritime disasters like the one portrayed in the film "The Perfect Storm." Code-named "Blue Wave," the new IBM RS/6000 SP will rank as the most powerful supercomputer at the Defense Department and the fourth-fastest in operation anywhere in

the world. It will enable the U.S. Navy to create the most detailed model of the world's oceans ever constructed. The research performed by "Blue Wave" is expected to improve maritime storm forecasting as well as search and rescue efforts for naval vessels. In June, Armonk, N.Y.-based IBM unveiled the fastest computer in the world, able to process more in a second than one person with a calculator could do in 10 million years. That supercomputer was designed for the U.S. government to simulate nuclear weapons tests. It was made for the Department of Energy's Accelerated Strategic Computing Initiative (ASCI). IBM sold the system, which occupies floor space equivalent to two basketball courts and weighs as much as 17 elephants, to the DOE for $110 million. The Navy computer, which can process two trillion calculations per second, will model ocean depth, temperature and wave heights to new levels of accuracy and detail, boosting the ability of meteorologists to predict storms at sea. "The Perfect Storm," a best-selling book by Sebastian Junger recently made into a film, told the tale of the Andrea Gail, a fishing vessel at sea off the coast of Newfoundland during a deadly storm that killed the entire crew In 1999, IBM became the leader in the traditional supercomputer market. IBM now has about 30 percent of that market, in which some 250 computers that range in price from $2 million to $100 million or more are sold every year, for use in weather predictions, research and encryption.

Sapphire Slams A Worm Into .Earth
Shatters All Previous Infection Rates

San Diego - Feb 04, 2003

A team of network security experts in California has determined that the computer worm that attacked and hobbled the global Internet 11 days ago was the fastest computer worm ever recorded. In a technical paper released Tuesday, the experts report that the speed and nature of the Sapphire worm (also called Slammer) represent significant and worrisome milestones in the evolution of computer worms. Computer scientists at the University of California, San Diego and its San Diego Supercomputer Center (SDSC), Eureka-based Silicon Defense, the University of California, Berkeley, and the nonprofit International Computer Science Institute in Berkeley, found that the Sapphire worm doubled its numbers every 8.5 seconds during the explosive first minute of its attack. Within 10 minutes of debuting at 5:30 a.m. (UTC) Jan. 25 (9:30 p.m. PST, Jan. 24) the worm was observed to have infected more than 75,000 vulnerable hosts. Thousands of other hosts may also have been infected worldwide. The infected hosts spewed billions of copies of the worm into cyberspace, significantly slowing Internet traffic, and interfering with many business services that rely on the Internet. "The Sapphire/Slammer worm represents a major new threat in computer worm technology, demonstrating that lightning-fast computer worms are not just a theoretical threat, but a reality," said Stuart Staniford, president and founder of Silicon Defense. "Although this particular computer worm did not carry a malicious payload, it did a lot of harm by spreading so aggressively and blocking networks."
taking out earth.com takes but a worm or two or three or more ......

The Sapphire worm's software instructions, at 376 bytes, are about the length of the text in this paragraph, or only one-tenth the size of the Code Red worm, which spread through the Internet in July 2001. Sapphire's tiny size enabled it to reproduce rapidly and also fit into a type of network "packet" that was sent one-way to potential victims, an aggressive approach designed to infect all vulnerable machines rapidly and saturate the Internet's bandwidth, the experts said. In comparison, the Code Red worm spread much more slowly not only because it took longer to replicate, but also because infected machines sent a different type of message to potential victims that required them to wait for responses before subsequently attacking other vulnerable machines. The Code Red worm ended up infecting 359,000 hosts, in contrast to the approximately 75,000 machines that Sapphire hit. However, Code Red took about 12 hours to do most of its dirty work, a snail's pace compared with the speedy Sapphire. The Code Red worm sent six copies of itself from each infected machine every second, in effect "scanning" the Internet randomly for vulnerable machines. In contrast, the speed with which the diminutive Sapphire worm copied itself and scanned the Internet for additional vulnerable hosts was limited only by the capacity of individual network connections. "For example, the Sapphire worm infecting a computer with a one-megabit-persecond connection is capable of sending out 300 copies of itself each second," said Staniford. A single computer with a 100-megabit-per-second connection, found at many universities and large corporations, would allow the worm to scan 30,000 machines per second.

"The novel feature of this worm, compared to all the other worms we've studied, is its incredible speed: it flooded the Internet with copies of itself so aggressively that it basically clogged the available bandwidth and interfered with its own growth," said David Moore, an Internet researcher at SDSC's Cooperative Association for Internet Data Analysis (CAIDA) and a Ph.D. candidate at UCSD under the direction of Stefan Savage, an assistant professor in the Department of Computer Science and Engineering. "Although our colleagues at Silicon Defense and UC Berkeley had predicted the possibility of such high-speed worms on theoretical grounds, Sapphire is the first such incredibly fast worm to be released by computer hackers into the wild," said Moore. Sapphire exploited a known vulnerability in Microsoft SQL servers used for database management, and MSDE 2000, a mini version of SQL for desktop use. Although Microsoft had made a patch available, many machines did not have the patch installed when Sapphire struck. Fortunately, even the successfully attacked machines were only temporarily out of service. "Sapphire's greatest harm was caused by collateral damage—a denial of legitimate service by taking database servers out of operation and overloading networks," said Colleen Shannon, a CAIDA researcher. "At Sapphire's peak, it was scanning 55 million hosts per second, causing a computer version of freeway gridlock when all the available lanes are bumper-tobumper." Many operators of infected computers shut down their machines, disconnected them from the Internet, installed the Microsoft patch, and turned them back on with few, if any, ill effects. The team in California investigating the attack relied on data gathered by an array of Internet "telescopes" strategically placed at network junctions around the

globe. These devices sampled billions of information-containing "packets" analogous to the way telescopes gather photons. With the Internet telescopes, the team found that nearly 43 percent of the machines that became infected are located in the United States, almost 12 percent are in South Korea, and more than 6 percent are in China. Despite the worm's success in wreaking temporary havoc, the technical report analyzing Sapphire states that the worm's designers made several "mistakes" that significantly reduced the worm's distribution capability. For example, the worm combined high-speed replication with a commonly used random number generator to send messages to every vulnerable server connected to the Internet. This so-called scanning behavior is much like a burglar randomly rattling doorknobs, looking for one that isn't locked. However, the authors made several mistakes in adapting the random number generator. Had not there been enough correct instructions to compensate for the mistakes, the errors would have prevented Sapphire from reaching large portions of the Internet. The analysis of the worm revealed no intent to harm its infected hosts. "If the authors of Sapphire had desired, they could have made a slightly larger version that could have erased the hard drives of infected machines," said Nicholas Weaver, a researcher in the Computer Science Department at UC Berkeley. "Thankfully, that didn't occur."

University of Hawaii will use new IBM supercomputer to investigate Earth's meteorological mysteries
"Blue Hawaii" system marks innovative partnership between university, Maui High Performance Computing Center and IBM

Honolulu, HI, October 25, 2000—The University of Hawaii (UH) today introduced an IBM supercomputer code-named "Blue Hawaii" that will explore the inner workings of active hurricanes, helping university researchers develop a greater understanding of the forces driving these destructive storms. The IBM SP system—the first supercomputer ever installed at the University of Hawaii—is the result of an initiative by the Maui High Performance Computing Center (MHPCC) in collaboration with IBM. This initiative has culminated in an innovative partnership between the university, MHPCC and IBM. "We're delighted to have IBM and MHPCC as partners," said university president Kenneth P. Mortimer. "This new supercomputer adds immeasurably to the technological capacity of our engineering and science programs and will propel us to a leadership position in weather research and prediction." Donated by IBM to the university, Blue Hawaii is the technological heir to IBM's Deep Blue supercomputer that defeated chess champion Garry Kasparov in 1997. Blue Hawaii will power a wide spectrum of University of Hawaii research efforts, such as:
 
Hurricane research. Wind velocity data acquired from weather balloons and aircraft-borne labs will be analyzed to develop a greater understanding of the forces that drive hurricanes. This will enhance meteorologists' ability to predict the storms. Climate modeling. Scientists will investigate the interaction between the oceans and the atmosphere believed to cause long-term climate variations. The research is expected to lead to a more accurate method for predicting changes in the world's climate, which will benefit numerous industrial sectors, including agriculture, manufacturing, and transportation. Weather forecasting. Meteorological data will be processed through state-of-the-art computer models to produce weather forecasts for each of Hawaii's counties.



In addition, scientists will rely on the supercomputer for a number of vital research projects in the areas of physics and chemistry. Educational programs in the university's Department of Information and Computer Sciences will also be

developed to train graduate students in computational science, which involves using high-performance computers for simulation in scientific research projects. "This supercomputer strengthens our reputation as a location with a burgeoning high technology industry," Hawaii Governor Benjamin Cayetano said. "It is an opportunity for our students and educators to work with a powerful research tool. This donation by IBM boosts this Administration's own support of the university's technology-related programs." The synergy between UH, MHPCC, and IBM will provide the resources needed to establish UH as a leader in research computing. MHPCC, an expert in production-level computing on the SP supercomputer, is acting as an advisor to UH on a broad range of technical topics and will install and prepare the supercomputer for UH. In addition, MHPCC and IBM will assist UH researchers in using the new research tool. Located in the Department of Information and Computer Sciences at the university's Pacific Ocean Science and Technology Building, Blue Hawaii is powered by 32 IBM POWER2 microprocessors, 16 gigabytes of memory and 493 gigabytes of IBM disk storage. The machine substantially augments the supercomputing power that's based in the state of Hawaii, already home to MHPCC, one of the world's most prestigious supercomputer facilities. Together, Blue Hawaii and MHPCC form a powerful technology foundation for the burgeoning scientific research initiatives located in Hawaii. In the past five years, government research grants awarded to Hawaii scientists have increased by 34 percent to $103 million, according to the UH office of research services. "Scientists at the University of Hawaii are conducting exciting research across a number of important disciplines," said IBM vice president Peter Ungaro. "IBM is proud to work with UH and MHPCC in providing the university with the industry's most popular supercomputer, which will help researchers achieve their important goals more quickly and with better results." Most Popular Supercomputer The Blue Hawaii system joins a long roster of IBM SP supercomputers around

the world. According to the TOP500 Supercomputer List*, IBM SPs now account for 144 of the world's 500 most powerful high performance computers—more than any other machine. The list is published twice a year by supercomputing experts Jack Dongarra from the University of Tennessee and Erich Strohmaier and Hans Meuer of the University of Mannheim (Germany). IBM SP supercomputers are used to solve the most complex scientific and business problems. With the IBM SP, scientists can model the effects of the forces exerted by galaxies; corporations can perform complex calculations on massive amounts of data in order to support business decisions; petroleum exploration companies can rapidly process seismic data to determine where they should drill; and company executives seeking to meet Internet demand can enable complex Web-based transactions. About the University of Hawaii The University of Hawaii is the state's 10-campus system of public higher education. The 17,000-student Manoa campus is a Carnegie I research university of international standing that offers an extensive array of undergraduate, graduate and professional degrees. The university's research program last year drew $179 million in extramural funding and is widely recognized for its strengths in tropical medicine, evolutionary biology, astronomy, oceanography, volcanology, geology and geophysics, tropical agriculture, electrical engineering and Asian and Pacific studies. Visit UH at www.hawaii.edu. About MHPCC MHPCC is ranked among the Top 100 most powerful supercomputer facilities in the world. MHPCC provides DoD, government, private industry, and academic users with access to leading edge, high performance technology. MHPCC is a center of the University of New Mexico established through a cooperative agreement with the U.S. Air Force Research Laboratory's Directed Energy Directorate. MHPCC is a Distributed Center of the DoD High Performance Computing Modernization Program (HPCMP), a SuperNode of the

National Science Foundation's National Computational Science Alliance, and a member of Hawaii's growing science and technology community.

Technology use in Super Computer

Pipelining
The most straightforward way to get more performance out of a processing unit is to speed up the clock (setting aside, for the moment, fully asynchronous designs, which one doesn't find in this space for a number of reasons). Some very early computers even had a knob to continuously adjust the clock rate to match the program being run. But there are, of course, physical limitations on the rate at which operations can be performed. The act of fetching, decoding, and executing instructions is rather complex, even for a deliberately simplified instruction set, and there is a lot of sequentiality. There will be some minimum number of sequential gates, and thus, for a given gate delay, a minimum execution time, T(emin). By saving intermediate results of substages of execution in latches, and clocking those latches as well as the CPU inputs/outputs, execution of multiple instructions can be overlapped. Total time for the execution of a single instruction is no less, and in fact will tend to be greater, than T(emin). But the rate of instruction execution, or issue rate, can be increased by a factor proportional to the number of pipe stages. The technique became practical in the mid-1960s. The Manchester Atlas and the IBM Stretch project were two of the first functioning pipelined processors. From the IBM 390/91 onward, all state-of-the-art scientific computers have been pipelined. Multiple Pipelines Not every instruction requires all of the resources of a CPU. In "classical" computers, instructions tend to fall into categories: those which perform memory operations, those which perform integer computations, those which operate on floating-point values, etc. It us thus not too difficult for the processor pipeline to thus be further broken down "horizontally" into pipelined functional units,

executing independently of one another. Fetch and decode are common to the execution of all instructions, however, and quickly become a bottleneck. Limits to Pipelining Once the operation of a CPU is pipelined, it is fairly easy for the clock rate of the CPU to vastly exceed the cycle rate of memory, starving the decode logic of instructions. Advanced main memory designs can ameliorate the problem, but there are always technological limits. One simple mechanism to leverage instruction bandwidth across a larger number of pipelines is SIMD (Single Instruction/Multiple Data) processing, wherein the same operation is performed across ordered collections of data. Vector processing is the SIMD paradigm that has seen the most visible success in high-performance computing, but the scalability of the model has also made it appealing for massively parallel designs. Another way to ameliorate the memory latency effects on instruction issue is to stage instructions in a temporary store closer to the processor's decode logic.

Instruction buffers are one such structure, filled from instruction memory in
advance of their being needed. An instruction cache is a larger and more persistent store, capable of holding a significant portion of a program across multiple iterations. With effective instruction cache technology, instruction fetch bandwidth has become much less of a limiting factor in CPU performance. This has pushed the bottleneck forward into the CPU logic common to all instructions: decode and issue. Superscalar design and VLIW architectures are the principal techniques in use today (1998) to attack that problem.

Scalable Vector Parallel Computers
The Scalable Vector Parallel Computer Architecture is an architecure where vector processing is combined with a scalable system design and software. The major components of this architecture are a vector processor as the single processing node, a scalable high performance interconnection network, including the scalability of I/O, and system software which supports parallel processing at

a level beyond loosely coupled network computing. The emergence of a new Japanese computer architecture comes to the surprise of many who are used to thinking that Japanese companies never undertake a radical departure from existing architectures. Nonetheless scalable vector parallel computers are an original Japanese development, which keeps the advantages of a powerful single processor but removes the restrictions of shared memory vector multiprocessing. The basic idea of this new architecture is to implement an existing vector processor in CMOS technology and build a scalable parallel computer out of these powerful single processors. The development is at the same time conservative and innovative in the sense that two successful and meanwhile proven supercomputer design principles, namely vector processing and scalable parallel processing, are combined to give a new computer architecture.

Vector Processing
Vector processing is intimately associated with the concept of a "supercomputer". As with most architectural techniques for achieving high performance, it exploits regularities in the structure of computation, in this case, the fact that many codes contain loops that range over linear arrays of data performing symmetric operations. The origins of vector architecure lay in trying to address the problem of instruction bandwidth. By the end of the 1960's, it was possible to build multiple pipelined functional units, but the fetch and decode of instructions from memory was too slow to permit them to be fully exploited. Applying a single instruction to

multiple data elements (SIMD) is one simple and logical way to leverage limited instruction bandwidth.

Latest Technology
Andromeda™ The latest technology from Super Computer, Inc., code named "Andromeda™" provides game developers a rich set of tools designed to make game implementation easier and lower the time to market. Other features are designed to allow hosting a game server and server administration easy. In today's competitive gaming market, it is not only important to get your game to the stores quickly, but keep it selling. All of Andromeda™'s features help you do just that. Master Browser Service (MBS) "Push" technology allows your servers to be accurately listed in a single repository accessable from in-game browsers and third party aplpications. Andromeda™'s MBS uses a propriatry protocol that allows for both forwards and backwards compatability so that applications that retrieve information from the master browser need to go through updates to continue to show your servers. Because Andromeda™'s MBS is located in the fastest data center in the world you know that players can retrieve server lists quickly Server Authentication Service (SAS) Player and server authentication services provided by Andromeda™'s SAS are as flexable as they are robust. You can store everything from basic authentication information to full player settings allowing players to use their preferred configuration--even if they are on a friend's computer. SAS also works with Andromeda™'s Pay-Per-Play Service and Games-On-Demand Service to control whether a player may join a server. Statistics Tracking Service (STS) Andromeda™'s STS service uses streaming to collect player and server statistics in real-time. Information is fed into the STS database and processed resulting in accurate player and server statistics. Dynamic Content Delivery System (DCDS)

In-game dialogs change. Layouts change. Why patch? Andromeda™'s DCDS system can deliver in-game content on the fly. Caching technology allows Andromeda™ to update only what has changed--including game resources. Deliver dynamic content, such as news, forums, and server rental interfaces without having to patch. Combined with Andomeda's Server Authentication Service and Pay-Per-Play Service, DCDS can also deliver an interface to players to allow them to add more time to their account without leaving the game. Remote Console Interface The remote console interface provides all the tools to allow server administrators to remotely control the server using a standard remote console language. CVAR Management Interface Another server administration tool, the CVAR Management Interface manages server settings and, combined with the Remote Console Interface, can restrict the ability for players to change certain CVARs using the remote console. Netcode Interface Solid netcode is the backbone of any multiplayer game. Highperformance, reliable interfaces allows game developers to reduce their time to market and concentrate on developing their game. Pay-Per-Play Service Andromeda™'s Pay-Per-Play service allows game publishers to charge players by the minute, hour, day, or month to play online. This technology is both secure and reliable. Games-On-Demand Service This service allows a game server to created on demand for whatever duration the ccustomer desires. Whether they want the server for a few hours to play a match, or the same time every week for practices, this solution is an inexpensive way to obtain a high-performance server that meets tournament regulations

inexpensively. Other Features Andromeda™'s tools integrate together. For example, your in-game browser can pull up a player summary and their player stats all through the MBS without having to connect to the SAS and STS. High availability clustering technology ensures that Andromeda™ is available 24/7/365.

ClanBuilder™ features total clan resource management including a roster, calendar, news board, gallery, links, remote game server console and robust security. The ClanBuilder™ roster is a fully customizable roster allowing guild masters to create their own fields, squads, ranks, and awards. The roster supports ICQ allowing visitors to add guild members to their ICQ contact list or send a message directly from the roster at the click of a button. The ClanBuilder™ calendar allows events to be created for the guild as a whole or a specific division. Recurrence patterns allow guild masters to easily create recurring events. ClanBuilder™ will even adjust the date and time of an event based on the time zone a vistor picks. The ClanBuilder™ news board is a powerful tool that allows news articles to be posted by any visitor requiring approval by the guild master before appearing on the news board. Like the rest of the ClanBuilder™ tools, division specific news can be posted. The ClanBuilder™ links and gallery allow screen-shots and links to be easily added and categorized. Images are stored automatically on the ClanBuilder™ server without the need for FTP or your own web space. The ClanBuilder™ game servers provides not only a place to list your clan game servers, but also allows remote console to be used with supported games. The ClanBuilder™ security model is simple yet powerful allowing guild masters to delegate administrative tasks to specific members and customize which members can make changes to any tool. ClanBuilder™ offers a rich set of tools allowing full customization

down to the individual colors on the screen. Easy to use, yet powerful, no clan can afford to be without this time-saving tool. Get into the game fast, with ClanBuilder™.

Architectural Themes
Looking over the numerous independent high-performance computer designs in the 1970's-90's, one can discern several themes or schools of thought which span a number of architectures. The following pages will try to put those in some kind of perspective. Contributions of additional links to seminal articles in these categories are more than welcome. Email to: KevinK@acm.org Pipelining Dealing with Memory Latency Vector Processing Parallel Processing Massive Parallelism Commoditization

SUPER COMPUTER SELECTS WorldCom FOR HOSTING SERVICES WorldCom, a the leading global business data and Internet communications provider, announced that Super Computer Inc, a revolutionary online computer gaming company, has selected WorldCom Internet Colocation Services to power its cutting-edge services which support millions of online video game players. Super Computer Inc SCI) was created for the 13 million online video game players to improve the quality of their experience while addressing the high cost of the game server rental industry. To establish a leadership position in the rapidly growing market, SCI developed a supercomputer capable of delivering game server access with unsurpassed speeds, reliability and value. By choosing WorldCom, SCI benefits from the cost and operating efficiencies of colocation outsourcing, while leveraging the performance, security and reliability of WorldCom's world-class Internet data centers and direct, scalable, high-speed connections to the facilities-based WorldCom global IP network. Financial terms of the agreement were not disclosed. "WorldCom has long been a tier one Internet provider, and unquestionably has the most superior and expansive IP network on the planet," said Jesper Jensen, president of Super Computer. "The two biggest factors for success in online gaming are network performance and equipment performance. We require access to the best network in order for us to be successful."

SCI's colocation solution ensures that its gaming servers will be up and running 24/7, even at peak usage times. With its industry-leading six-point Internet colocation performance guarantees, WorldCom assures 100% network availability, 100 percent power availability and 99.5% packet delivery, as well as other key network performance metrics. In addition, Super Computer chose WorldCom because its network could easily scale and provide the necessary bandwidth - beyond speeds of 1.6 Gbps - as it grows. "The Super Computer agreement highlights the fact that WorldCom is a viable provider today and in the future," said Rebecca Carr, WorldCom director of Global Hosting Services. "By outsourcing to WorldCom, SCI can leverage our Internet expertise and our global Internet footprint to deliver the world- class speed and reliability of our network to its customers." Currently colocated in WorldCom's state-of-the-art Atlanta data center, the company plans to expand into additional centers around the globe over the next nine months. With the support of its managed Web and application hosting affiliate Digex, WorldCom offers the full array of Web hosting solutions from colocation to shared, dedicated and custom managed hosting. Each runs over WorldCom's facilities-based global network, through WorldCom's state-of-the-art data centers across the U.S., Canada, Europe and Asia Pacific. Web hosting services are part of the full continuum of advanced data communications solutions WorldCom provides to business customers around the world.

About Super Computer Inc
Super Computer Inc (SCI) is one of the world's fastest growing game hosting solutions. With the invention of the world's first supercomputer for FPS game hosting, the Jupiter Cluster, and the Callisto mini-cluster for Broadband Providers, SCI is working with the gaming industry to consolidate the game hosting market. With the move to WorldCom's network, SCI now offers the world's fastest gaming technology and connectivity. SCI introduced the concept of games hosted on supercomputers to the consumers in June 2002.

About WorldCom Inc
WorldCom Inc is a pre-eminent global communications provider for the digital generation, operating in more than 65 countries. With one of the most expansive, wholly-owned IP networks in the world, WorldCom provides innovative data and Internet services for businesses to communicate in today's market. In April 2002, WorldCom launched The Neighborhood built by MCI -- the industry's first truly any-distance, all-inclusive local and long-distance offering to consumers.


								
To top