HPCwire: Sequoia: The Next Generation of Supercomputer

The Leading Source for Global News and Information Covering the Ecosystem of High Productivity Computing

HPCwire >> Features

April 28, 2009

Sequoia: The Next Generation of Supercomputer

by Markus Henkel

Page: 1 of 2
1 | 2 All »

What would the former IBM chief executive Thomas Watson have to say about the current development taking place with computers? After all it was Watson who claimed in 1943 that the world would need about five computers. Today, that claim looks a little wayward, to say the least, with millions of computers being sold over-the-counter every year.

When computer performance is bundled together, we have what is known as a supercomputer. In the US, plans are already underway to build just such a computing colossus that will overshadow all previous such systems. The "Sequoia" project aims to build the first computer capable of reaching the 20 petaflop mark. In comparison, "Roadrunner," currently the world's fastest computer, just about manages a petaflop, meaning that "Sequoia" would make it look more like the world's fastest calculator.

But what does 20 petaflops per second mean? Primarily, 20 petaflops is a value, a 20 with fifteen zeros, and is in itself nothing tangible. In order to form a better appreciation for what "Sequoia" with its 20 petaflops is capable of, it would take six billion people all equipped with calculators 1,000 years to do the calculations that "Sequoia" can manage in one day. That's a scenario that in Watson's day would have caused an uproar and been considered as being beyond even the wildest borders of fiction.

Before "Sequoia" is put into operation, another system will ensure that everything runs exactly to plan. The supercomputer "Dawn," primarily a delivery system, will be based on Blue Gene/P technology and reach performances of over 500 teraflops. Both computers will work in tandem, although "Dawn" will afford users the opportunity of developing or adapting their applications for Blue Gene technology and to test and improve their scalability. "Dawn" is, as such, a typical porting and developing system. It will be the system on which applications are created and these applications will then execute operations and calculations in the petaflop range on "Seqouia." Since there are not so many of these applications around, the supposedly smaller computer takes on added significance for users. They can undertake and carry out initial tests and studies and attempt to pave the way toward such petaflop applications.

The National Nuclear Security Administration, which commissioned the project, is a part of the US Department of Energy. It wants to see "Sequoia" in use by 2012. By then no fewer than 96 racks will provide accommodation for the 1.6 million IBM POWER processors. Klaus Gottschalk
According to official press releases, "Sequoia" will contribute to increased security and reliability of the United States nuclear arsenal. It goes without saying, of course, that other types of security aspects pertaining to the nuclear arsenal will be simulated, especially with regard to keeping a secure eye on aging materials. All over the world, scientists have been searching for solutions to problems raised by the safe disposal and storage of nuclear waste.

"We see the entire project from the point of view of the researcher," said Klaus Gottschalk, IT Systems architect with IBM. "For him the use of the computer is easy to evaluate. Large sums are being invested to help drive development onwards."

However, this giant machine is not only capable of turning nuclear research into visible, viewable action. The enormous potential offered by a 20 petaflop computer extends to far beyond nuclear weapons safety. According to IBM estimates, the supercomputer will be able to forecast weather up to 40 times more precisely than is possible today, and be invaluable in such areas as astronomy, energy, biotechnology and climate research. "Modeling and simulation is crucial for ensuring the ability of our country to innovate and compete globally," explained Dr. Cynthia McIntyre, Senior VP at the Council on Competitiveness.

At this point, IBM has not said exactly how much power "Sequoia" is going to need. But according to the company, the machine is set to break all records in this area as well. It has been estimated that it will be the world's first computer to achieve an efficiency of 3,050 calculations per watt.

In terms of supercomputing, the US is no longer the only big player. The IBM-JUGENE system in Juelich, Germany, means Europe is currently ranked 11th in a list of the world's 500 fastest computers compiled by the universities of Mannheim and Tennessee. Accordingly, the Juelich Research Center has been top of the tree in Europe for the last two years in terms of fastest computer. Plans are already afoot in Juelich to install the first petaflop computer in Europe -- incidentally also from IBM -- by the middle of this year.

In all probability, after an initial introduction, this supercomputer will force its way into the top three of the world's fastest computers. It will be capable of one quadrillion computational operations per second. The new supercomputer's roughly 295,000 processors will then be housed in 72 phonebox-sized cabinets in the computing labs of the Juelich Supercomputing Center. Replete with 144 terabytes of RAM, and together with the remaining computers at the research center, Juelich will then be operating at 1.3 petaflops per second. In addition to its high speed, the supercomputer will also have access to around 6 petabytes of hard disk. That more or less corresponds to sufficient memory to store all the information contained on over one million DVDs.

Page: 1 of 2
1 | 2 All »

Article Tools

Share Options

(Digg, Technorati, more)

Subscribe

Subscribe to HPCwire

Discussion

There are 0 discussion items posted.

Join the Discussion

You must be a registered user in order to comment on HPCwire stories.

Members enjoy:

	Access to Special Content, including Podcasts, Webcasts, and Whitepapers and more!
	Ability to take an active role in the discussions that are shaping the HPC technology discipline!
	Special offers available only to HPCwire members!

Become a Registered User Today!

Registered Users Log in Here to Comment

Email Address:	Password (case sensitive)

Remember me	Forgot Password?

RSS Feeds

Feeds by Topic

Feeds by Industry

Feeds by Content Type

Top Headlines

IBM Releases Energy Efficient Power7 System

Feb 09 | eWeek Europe | Company says new high-end servers will deliver "intelligent performance." Read more...

Inductive Coupling Packs Flash Drive in a Chip

Feb 09 | EE Times | Wireless technology promises energy-efficient chip-to-chip communication. Read more...

IBM, Microsoft Help Create Montana Supercomputer

Feb 08 | eWeek | A new kind of Rocky Mountain high. Read more...

AMD Aims for GPUs in Mainstream Servers Starting 2012

Feb 08 | Computerworld | Chip maker hopes to bring CPU-GPU processors to servers in two years. Read more...

Graphene Transistors That Work at Blistering Speeds

Feb 05 | Technology Review | IBM has created graphene transistors that leave silicon ones in the dust. Read more...

Read more headlines...

Featured Whitepapers

Virtualization for Aggregation And The vSMP Architecture™

Jan 12 | | In-depth look at vSMP Foundation server virtualization technology, technical implementation, use cases and capabilities. The technical whitepaper provides an architectural overview and details on the three vSMP Foundation products: vSMP Foundation for SMP, vSMP Foundation for Cluster and vSMP Foundation for Cloud.

Copper Cable Technologies for High Performance Computing

Jan 18 | | This white paper discusses Gore’s copper cable assemblies, and how they continue to exceed the standards for providing reliable, cost-effective solutions for high-performance computer applications.

Appro Assists LLNL with Cluster Designed for Extreme Scale Visualization

Jan 11 | | LLNL is home to some of the fastest computers in the world. In 2012, LLNL expects to have the Sequoia supercomputing cluster operational with a projected performance of over 20 PFLOP/s. These systems will focus on strengthening the foundations of predictive simulation through running large suites of complex simulations and then comparing model predictions with experimental data. To visualize this project’s large amount of data, LLNL requested an Appro Supercomputing Cluster specifically designed to support interactive data analysis.

View the White Paper Library

Multimedia

Webcast: Virtualized Data Center Roundtable

Join this online panel discussion for live Q&A with leading industry experts, analysts, and end-users to discuss the latest innovations, best practices, barriers to implementation, and measurable benefits of server virtualization with a particular focus on today's real world solutions.

Webcast: Watch SC09 Birds of a Feather Video: Scalable Fault-Tolerant HPC Supercomputers

Learn about scalable fault-tolerant architectures and examples of energy efficient and scalable supercomputing clusters using dual QDR InfiniBand to combine capacity computing with network failover capabilities with the help of programming languages such as MPI and a robust Linux cluster management package.

Webcast: High Performance Computing for a Smarter Planet

LIVE@SCO9: The IBM team discusses new innovations in hardware, software and services that help clients better understand their workloads and get insight from their R&D efforts. Technology demonstrations include the soon-to-be-released Power7 HPC processor, the DCS990 system with 2.4 petabytes of storage, the xCAT management tool, secure HPC cloud computing and more. Winners of two HPCwire Readers' and Editors’ Choice Awards! Take the IBM virtual tour at SC09 or more information go online to: http://www-03.ibm.com/systems/deepcomputing/sc09.html