Georgia State Adds New IBM Supercomputer to Keep Pace With Demand

Georgia State University and IBM (NYSE: IBM) announced today that the university has purchased an IBM System Cluster 1350 supercomputer through a regional partnership program between IBM and the Southeastern Universities Research Association (SURA). This purchase is the latest addition to the school's expanding inventory of supercomputing resources.

Georgia State is building a diverse and powerful supercomputing center to support a growing body of academic and scientific research. In September 2006 three researchers began working with a new IBM System p5 575 with Power5+ processors running AIX 5.3 -- acquired as part of the IBM-SURA regional partnership. Those three users became a dozen in January 2007 and now include more than 40 faculty from many disciplines as well as from other SURA sites such as Virginia Commonwealth University, Louisiana State University and University of Delaware.

"We're experiencing increasing demand for supercomputer resources," said Art Vandenberg, director of the university's supercomputing center. "Our strategy is to build a center that supports a wide variety of disciplines. So, we're adding the Cluster 1350 to support research that requires commodity level, Linux computing."

The new IBM System Cluster 1350 will help meet this increasing demand. As researchers gain access to more capacity and capability, they naturally are expanding their goals, driving demand. Researchers in chemistry, biology, molecular dynamics, computer science, climate modeling and political science will be among the new machine's first users. The Cluster 1350 will be added to Georgia State's high-end IBM System p5 575 supercomputer to offer scientists a range of in-house computing capabilities.

The 320 Intel® Xeon® Quad-Core processors that make up Georgia State's new Cluster 1350 will also be shared by SURA -- a group of 60 schools that shares supercomputing resources via a grid setup known as SURAgrid. The university's Cluster 1350 configuration runs at approximately three teraflops, or three billion calculations per second.

"Georgia State University has been a long term contributor to the SURAgrid initiative and we are extremely pleased that SURA has been able to assist Georgia State and other SURAgrid participants by providing access to specially discounted hardware and services through our partnership with IBM," said Gary Crane, SURA Director of IT Initiatives.

Parasites & Politics
A partial list of the university's research projects that will benefit from the new IBM supercomputer includes:

Dr. Tarynn Witten, Virginia Commonwealth University, plans to create more complex and biologically realistic models to investigate Chagas' Disease. The long-term goal of this project is to apply novel mathematical and computational modeling technologies to develop new paradigms for understanding the infectious disease process. (http://www.vcu.edu/csbc/vpp)
Maryam Rahimian, PhD student in chemistry at Georgia State University, also plans to study parasitic diseases such as sleeping sickness, malaria and leishmania. The goal is to find new drugs with high efficiency and low cost. The IBM HPC Systems enable Rahimian to run more complex models which can provide valuable information on the influence of solvent and molecular motion in DNA complexes.
Eric Hurst, PhD student in political science, intends to apply the university's additional compute power to one of the first political science studies to run on a supercomputer. Hurst intends to study the complete voting history of independent regulatory commissions, such as the SEC, to help determine how ideology and bureaucracy affect the nation's policies and politics.
The IBM System Cluster 1350 will be configured with additional IBM solutions including: BladeCenter® H technology, HS21 XM Blade servers utilizing two Quad-Core Intel Xeon processors, 10G Ethernet backbone, IBM's xCAT and Loadleveler cluster management tools. It will also have access to existing Tivoli Storage Manager and 18 Terabytes of storage that Georgia State has acquired during the past year. This will allow researchers to easily share results from the IBM System p5 with the new IBM System Cluster 1350 -- and vice versa.

Georgia State University
Located in the heart of downtown Atlanta, Georgia State University is one of the leading research institutions in an urban setting. The university offers 52 undergraduate and graduate degree programs in 250 fields of study through six college-level units. Preparing students for lifelong learning in a global society, Georgia State's degree programs are relevant to the practical needs of both the student and the community. A member of the University System of Georgia, Georgia State is the USG's most diverse institution, with more than 50,000 students enrolled annually from every county in Georgia, every state in the nation and 159 countries. For more information, visit www.gsu.edu.

The Southeastern Universities Research Association (SURA) is a consortium of over 60 leading research institutions in the southern United States and the District of Columbia established in 1980 as a non-stock, nonprofit corporation. SURA serves as an entity through which colleges, universities, and other organizations may cooperate with one another, and with government and industry in acquiring, developing, and using laboratories and other research facilities and in furthering knowledge and the application of that knowledge in the physical, biological, and other natural sciences and engineering. For more information, visit www.sura.org.

About SURAgrid
SURAgrid is a consortium of organizations collaborating and combining high performance computing and advanced networking resources to improve the research and learning infrastructure of the SURA region. The vision for SURAgrid is to orchestrate access to a rich set of cyberinfrastructure capabilities to meet diverse needs of university researchers and educators. Capabilities being cultivated include locally contributed high performance computing resources, project-specific tools and environments, shared access to highly specialized high performance computing and cyberinfrastructure resources, and gateways to national and international cyberinfrastructure. For more information on SURAgrid see: www.sura.org/suragrid

IBM System Cluster 1350
The IBM System Cluster 1350 is designed for a broad range of supercomputer application environments, including industrial design and manufacturing, financial services, life sciences, government and education. The IBM System Cluster 1350 leverages IBM's extensive supercomputing experience to help minimize complexity and risk. Using advanced Intel® Xeon®, AMD Opteron™, Cell Broadband Engine and IBM PowerPC® processor-based server nodes, proven cluster management software and optional high-speed interconnects, the Cluster 1350 offers the best of IBM and third-party technology. For more information visit http://www-03.ibm.com/systems/clusters/hardware/1350.html

New 'Adria' Supercomputer Among 250 Most Powerful in the World

Advancing development of hydropower as a green alternative to traditional energy sources, world-renowned hydro-energy research organisation Turboinštitut has turned to IBM (NYSE: IBM) to deliver the most powerful supercomputer in South East Europe.

Named 'Adria' and based at the newly-established Ljubljana Supercomputing Center in Slovenia, the Linux-based system will be used by Turboinštitut to carry out complex water turbine design simulations in fractions of existing timescales. Performing calculations 50 times faster than current systems, Adria will reduce time-to-result from weeks to hours.

"By applying our supercomputer expertise and industry leadership, IBM will be fundamental to advances in Turboinštitut's hydro energy development programmes," said Biljana Weber, Country General Manager, IBM Slovenia. "IBM is increasingly mobilising its research, technology and insight to help clients worldwide address complex energy and environmental challenges."

The system will be instrumental in accelerating Turboinštitut's hydro-power research and development programme, encouraging energy companies worldwide to introduce the energy source as an environmentally friendly alternative.

"We selected IBM because of their expertise and flexibility. They quickly provided reference architecture and actively worked with our experts on specific areas of supercomputing architecture. IBM put in place a local team who will bring Adria to life and also ensured the financing of the investment," explained Dr. Vladimir Kercan, Executive Director, Turboinštitut.

The 2.048-core system uses 256 IBM BladeCenter servers, each carrying two quad-core processors with 4100 GB of RAM, and is expected to become one of the world's fastest Linux clusters (based on Top500.org). 'Adria' utilises the latest IBM solutions that reduce its own energy consumption.

The installation will be performed by IBM Slovenia and business partner SIMT.

For more information about IBM Slovenia, visit http://ibm.com/si

IBM Blue Genes Dominate List of Fastest Supercomputers

IBM's (NYSE: IBM) Blue Gene®/L supercomputer continued its four-year reign atop the TOP500 Supercomputer Sites list announced today. IBM systems account for 42% of the combined computational might of the planet's most powerful computers -- more than any other supercomputer vendor, according to the list.

In addition, the development version of IBM's newest Blue Gene/P solution(1) burst onto the list at number 30. Launched yesterday, the two-rack test system delivers 20.86 teraflops in a space the size of two refrigerator-sized racks. When it is fully scaled out, Blue Gene/P can be configured to reach up to three petaflops -- or three thousand trillion calculations per second.

IBM dominates the Top 10 with a total of six systems -- four of them Blue Gene supercomputers. The number one Blue Gene/L system, installed at Lawrence Livermore National Laboratory in California, delivers a sustained performance of 280.6 trillion operations per second or teraflops.

IBM systems account for 46 of the Top 100. Thirty-nine of them are built on IBM's POWER® architecture hardware running versions of Unix (AIX) and Linux software.

This year, two new IBM Blue Gene supercomputers entered into the Top 10. "New York Blue," at the New York Center for Computational Science in Stony Brook, N.Y., and the Blue Gene system at the Computational Center for Nanotechnology Innovations at Rensselaer Polytechnic Institute in Troy, N.Y.

Joining the Blue Gene/L system at Lawrence Livermore in the TOP500 list's Top 10 are IBM Watson Research Lab's own Blue Gene system at 91.29 teraflops. In addition to Blue Gene systems, IBM's POWER-based ASC Purple supercomputer at Lawrence Livermore National Laboratory is number six.

IBM's MareNostrum supercomputer, built on IBM BladeCenter JS21 blade servers and located at the Barcelona Supercomputing Center, retained its position as the most powerful supercomputer in Europe, ranking number 9 on the overall list. IBM also has the biggest system in China, a BladeCenter cluster at Sinopec ShengLI Oilfield Branch Company, Geophysical Research Institute.

Based on IBM's POWER Architecture, the IBM System Blue Gene Solution is optimized for bandwidth, scalability and the ability to handle large amounts of data while consuming a fraction of the power and floor space required by today's fastest systems. A variety of industries are using Blue Gene systems to advance their research capabilities for life sciences, financial modeling, hydrodynamics, quantum chemistry, molecular dynamics, astronomy and space research and climate modeling.

The "TOP500 Supercomputer Sites" is compiled and published by supercomputing experts Jack Dongarra from the University of Tennessee, Erich Strohmaier and Horst Simon of NERSC/Lawrence Berkeley National Laboratory and Hans Meuer of the University of Mannheim (Germany). The entire list can be viewed at www.top500.org

IBM Supercomputing Simulations Support Chip Breakthrough

IBM (NYSE: IBM) researchers today announced an advancement in computer-based simulations that is helping to drive chip technologies to new heights of performance and function. As reported in the scientific journal Physical Review Letters, a team of scientists at IBM's Zurich Research Laboratory for the first time used advanced supercomputer-based models to more deeply understand and master the complex behavior of a promising new material -- hafnium dioxide -- in silicon transistors, the fundamental building blocks of computer chips.

The new material is key to the company's recently-announced "high-k metal gate" technology, the first major change to the transistor since the emergence of silicon semiconductors, promising enhanced chip performance to benefit computers and other electronic systems. IBM is implementing the technology and will apply it to products in 2008.

The semiconductor industry has long sought to find a new material for a crucial part of the transistor known as the gate dielectric, which, with the materials used currently, is limiting the industry's ability to keep pace with the progress predicted by Moore's Law -- a maxim predicting a doubling of the number of transistors on a chip, and an associated increase in chip performance, every 12-18 months.

While hafnium dioxide appeared to be an ideal candidate for next-generation transistor gates, the introduction of any new material in semiconductors can have unforeseen consequences, so it must be thoroughly understood beforehand. One critical factor contributing to IBM's success in the highly complex and difficult task of integrating these new materials has been simulation of the interaction of this material at the atomic level.

Scientists at IBM's Zurich Research Laboratory have used their long-standing expertise in computer-based modeling, together with the capabilities offered by the IBM Blue Gene supercomputer, to determine why hafnium dioxide works so much better than other high-k materials previously considered by the industry. As a result, the researchers were able to gain a clear picture -- for the first time -- of the underlying physics driving the unique electrical behavior of hafnium dioxide when it mixes with silicon, as observed in laboratory experiments, shedding light on the reasons that make this material unique as gate dielectric.

For this study, the IBM team simulated various material compositions using 50 different models of hafnium silicates, materials that form when silicon and hafnium oxides mix. These models contain up to 600 atoms and approx. 5,000 electrons, representing a realistic system. A single calculation of the dielectric constant was accomplished in only five days of computing time on the two-rack Blue Gene/L supercomputer (4096 processors) installed at the Zurich lab's site. The complete simulation for all 50 models, approx. 250 days on Blue Gene, would normally take the most powerful laptop PC an astounding 700 years to calculate. This corresponds to a staggering 200 billion billion (2 x 10^20) operations.

This work illustrates how supercomputing modeling techniques are bringing technologies of all kinds to a new level. Computer simulations have been possible since the 1980s, but only now -- thanks to the evolution of algorithms and their mapping to such an extremely scalable and balanced architecture as that of the Blue Gene computer -- scientists are able to tackle these types of problems by creating realistic models with several thousands of particles, starting only from the laws of nature.

"Advances in algorithms and their optimal mapping on extremely large and scalable computer hardware such as Blue Gene are empowering us to do accurate and realistic atomic simulations of complex materials," explains Alessandro Curioni, supercomputing expert from IBM's Zurich lab and emphasizes: "So indeed, today we are able to use supercomputers to investigate materials that will be eventually used in the next generation of supercomputers."

The approach used by the IBM team is called ab initio molecular dynamics, where the interactions between the particles of the system are derived from the basic laws of physics without employing any empirical data. In the course of their ground-breaking work, the IBM team created more than 50 realistic virtual models of the hafnium silicates with various concentrations of hafnium on the supercomputer. They then simulated the evolution of these structures over a given time period, estimated their dielectric constants and used these results to rationalize experimental findings.

The advantage of computer-based simulations is that, being virtual, they are free of the problems inherent to laboratory experiments, such as the effects of preparation conditions, the purity of the compounds, or the presence of parasitic reactions. Most importantly, with the simulations one can follow what the individual atoms are doing. Computer simulations allow the "intrinsic" and ideal characteristics of a material to be calculated and correlated directly with the structure at an atomic level.

About the IBM Zurich Research Laboratory

The IBM Zurich Research Laboratory (ZRL) is the European branch of IBM Research. This network of some 3200 employees in eight laboratories around the globe is the largest industrial research organization in the world. ZRL itself currently employs about 320 people, representing more than 30 nationalities.

ZRL's Computational Biochemistry and Materials Science group is among the worldwide leading research teams in this field, and is active in the development of accurate and efficient methodologies for the simulation of diverse materials, ranging from silicon to enzymes, at the atomic level. Since August 2005, the team uses a two-rack Blue Gene/L system, with a peak performance of 11.2 Teraflops, installed on the ZRL site, making it the second fastest supercomputer in Switzerland.

www.zurich.ibm.com/deepcomputing