October 01, 2008
Last week, IBM and King Abdullah University of Science and Technology (KAUST) announced a collaboration to build "Shaheen," a 222 teraflop Blue Gene/P supercomputer. When deployed in 2009, it will represent the most powerful computer in the Middle East and one of the top systems in the world. Shaheen is intended to serve KAUST scientists, but will also be shared with a number of other institutions around the world.
Although the 16-rack Blue Gene/P will give the Saudis access to terascale computing next year, KAUST plans to install a petascale machine within two years, leading up to an exascale system when they become available. As such, the new supercomputer appears to be a significant step for Saudi Arabia in becoming a center for scientific research and the larger goal of transforming the country into an information society.
One of the shortest paths to becoming a global R&D player is through the acquisition of supercomputing infrastructure and science expertise, since these are key elements in developing knowledge-based industries, such as energy resource development, bioinformatics, nanotechnology and computational science. All you need is money, and Saudi Arabia has plenty of that. A multi-hundred teraflop machine, such as Shaheen will put the Saudis on par with the top systems in the U.S. and Europe, with the exception, perhaps, of a handful of petascale machines that will be deployed over the next year.
On the expertise side, teaming with IBM and a number of other institutions will enable KAUST to tap into a much larger reservoir of talent. Besides the partners mentioned in the press release -- IBM, MIT, London's Imperial College and the Hong Kong University of Science and Technology -- KAUST will also be collaborating with:
We got an opportunity to ask Majid Al-Ghaslan, KAUST's interim chief information officer (and the leader in the acquisition, design and development of the Shaheen supercomputer), a few questions about the upcoming system and how it would be used.
HPCwire: How is King Abdullah University of Science and Technology (KAUST) funded?
Al-Ghaslan: The University will be supported through a generous endowment established by King Abdullah. This multi-billion dollar fund will be among the largest university endowments in the world.
HPCwire: Can you disclose how much money KAUST has invested in "Shaheen"?
Al-Ghaslan: Because of contractual arrangements between KAUST and IBM, we are not at liberty to disclose the total cost of the supercomputer project. Having said that, the most important thing to realize is that KAUST will always make available the finest tools and world-class facilities, central amongst these is scientific computation capability.
HPCwire: Did you consider any other supercomputing architectures for the KAUST system besides Blue Gene/P? What were the deciding factors that led you to select the IBM system?
Al-Ghaslan: Several architectures were considered. It was clear that no other company could offer the end-to-end solutions like IBM. Some of the decisive features for us included a hugely-scalable and highly-reliable scientific computing platform while keeping power usage and space requirements to the minimum, and "right now" that's IBM! In addition, IBM was willing to develop a partnership from the bottom up with a fully-integrated approach in the region. KAUST is tapping into IBM's capability and dominance in the landscape of supercomputing.
HPCwire: Is there going to be any research supported by the new system that is unique to KAUST or its partners?
Al-Ghaslan: KAUST will house one of the most powerful supercomputers exclusively available to academic researchers and scientists where they will focus on four specific research thrusts: resources, energy and environment; biosciences and bioengineering; materials science and engineering; and applied mathematics and computational science. The "uniqueness" will probably come from data sets that are distinctive to Saudi Arabia -- Red Sea data, the Oil & Gas fields of Saudi Arabia, etc.
HPCwire: Also, are there any categories of commercial supercomputing applications that KAUST will focus on -- for example, oil and gas, financial analytics, bioinformatics, etc.?
Al-Ghaslan: KAUST will embark upon several research programs for its faculty and students with the goal to develop advanced scientific knowledge. To advance KAUST's mission and development as an institution, KAUST and The Global Research Partnership will contribute to the larger global research enterprise in areas as diverse as:
HPCwire: What's the strategy behind the research partnerships?
Al-Ghaslan: KAUST is an independent academic graduate-level research university. To date, KAUST has initiated several research partnerships which will provide KAUST with the ability to grow its high-quality research actively on campus. The partnership is one of KAUST's mechanisms for initiating research programs for its faculty and students, developing advanced scientific knowledge, advancing KAUST's mission and development as an institution, and contributing to the larger global research enterprise.
HPCwire: I assume the datacenter being built for Shaheen will have the capability to house an even larger system -- you talked about petaflop and exaflop systems. Can you give us an idea of how big the center will be and how much power will be supplied to it?
Al-Ghaslan: First you have to remember this is Saudi Arabia, energy is cheap here! KAUST will also have plenty of space given the campus total area, which spans more than 36 million square meters. We have future proofed the entire site with enough scaling for power and cooling capacity. The current state-of-the-art scientific datacenter being built has a total capacity of 500 racks and is housed in the same multi-story building where the relevant researchers and students will have their offices.
10/30/2013 | Cray, DDN, Mellanox, NetApp, ScaleMP, Supermicro, Xyratex | Creating data is easy… the challenge is getting it to the right place to make use of it. This paper discusses fresh solutions that can directly increase I/O efficiency, and the applications of these solutions to current, and new technology infrastructures.
10/01/2013 | IBM | A new trend is developing in the HPC space that is also affecting enterprise computing productivity with the arrival of “ultra-dense” hyper-scale servers.
Ken Claffey, SVP and General Manager at Xyratex, presents ClusterStor at the Vendor Showdown at ISC13 in Leipzig, Germany.
Join HPCwire Editor Nicole Hemsoth and Dr. David Bader from Georgia Tech as they take center stage on opening night at Atlanta's first Big Data Kick Off Week, filmed in front of a live audience. Nicole and David look at the evolution of HPC, today's big data challenges, discuss real world solutions, and reveal their predictions. Exactly what does the future holds for HPC?