HPCwire: TACC Embarks on the Path to Petascale

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

Leading HPC
Solution Providers

HPCwire >> Features

February 29, 2008

TACC Embarks on the Path to Petascale

by John Leidel, for HPCwire

I was so privileged as to make the pilgrimage down to the J.J. Pickle Research Center at the University of Texas for the dedication ceremony of Ranger, TACC's latest behemoth supercomputer. The list of attendees was surprisingly quite diverse. Principle investigators from a host of various disciplines, HPC heavyweights from the petroleum industry and several other big names in HPC all made appearances. Dr. Dan Atkins, director of the Office of Cyberinfrastructure, was there to see what his $59 million bought the NSF. Hector Ruiz, CEO of AMD, was there caressing what has become the largest single repository of quad-core Barcelona silicon in the world. One note of interest was Sun Microsystem's dedication to the event. Jonathan Schwartz, Scott McNealy, John Fowler and Andy Bechtolsheim [wearing his characteristic sandals] all made the trip to Austin. It was certainly a feat in itself to gather individuals of this magnitude in a common venue.

Despite the long list of industry heavyweights, all eyes seemed to be on Dr. Jay Boisseau, director of the Texas Advanced Computing Center. Jay's attitude at the event was analogous to that of a new father. There wasn't a single moment when you couldn't find Jay smiling ear-to-ear over the triumphant arrival of his silicon child. Ranger will certainly require the care and feeding typical of a newborn child. The 3,936 Sun blade modules emit a near-deafening whirl of noise. The power distribution units command more floor space than many large HPC installations and lifting the bundles of InfiniBand cabling would make Arnold Schwarzenegger strain. Needless to say, I was a kid in a candy store.

Much of the content delivered during the ceremony was directed toward the real meaning of "open science computing." Specifically, everyone discussed the impact of Ranger standing as the largest system in the world built for open science. I believe Dr. Atkins said it best. Ranger will be dedicated toward "finding knowledge needles in enormous data haystacks." How will they do so? I suggest we look at the numbers.

The Ranger system allocations will be split three ways. 90 percent of the allocations are headed for the TeraGrid. 5 percent are reserved for UT researchers and 5 percent are reserved for TACC's STAR program. This translates to ~56,678 cores dedicated to TeraGrid-allocated research projects. As such, 1.3 million CPU hours per day are directed toward open scientific research. TACC and the TeraGrid already have more than 500 users with allocations totaling over 100 million hours for the first quarter of operation. Those involved fully expect to allocate all 125 million hours of TeraGrid use for the second quarter of production. Over its total lifespan, Ranger will deliver more than 200,000 years of computational capability.

Ranger is most certainly the triumphant result of a successful partnership between vendor and end user. As we embark upon the age of petascale computing, it is imperative that vendors and end user organizations continue to form meaningful relationships in order to ease the pain from constructing and operating systems at such scale. I congratulate TACC, AMD and Sun Microsystems for their tireless efforts to provide such a large resource dedicated to facilitating open scientific research.

For more information on TACC and Ranger, point your browser to www.tacc.utexas.edu.

Only in Texas will you find steer horns mounted on a supercomputer

About the Author

John Leidel is a senior HPC architect and technical lead at a federal supercomputing laboratory and a chronic HPC geek. Having started his career as a classical software engineer, he participates in several open source projects centered on large enterprise computing/HPC environments. His recent interests include general purpose offload computing [FPGA/GPU/etc] in beowulf clusters. When he's not writing for insideHPC.com or slinging code for various open source pursuits, you can find him spending time with his family or whittling away at homemade briar pipes.

Article Tools

Share & Save Options

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

25 Years of Conventional Evaluation of Data Analysis Proves Worthless In Practice

Sep 05 | Uppsala University | Swedish researchers are revealing that "intelligent" computer-based methods for classifying patient samples is worthless when it comes to practical problems. Read more...

Three Dimensional Scans Could Revolutionise Brain Surgery

Sep 03 | Telegraph.co.uk | A new form of three dimensional scans could revolutionise brain surgery within a year, doctors claim. Read more...

Marching Penguins: Monitoring Your HPC Cluster

Sep 03 | Linux Magazine | In HPC, most attention is paid to utilization and performance, rather than service availability and problem notification. This article focuses on the latter Read more...

Big Data: Welcome to the Petacentre

Sep 03 | Nature News | What does it take to store bytes by the tens of thousands of trillions? Read more...

UD Scientist Finds Quicker Computer On-Ramp

Sep 01 | Delaware Online | In the world of supercomputer-powered science, speed is everything, and an open road can lead to the promised land. Read more...

Read more headlines...

Featured Whitepapers

SUSE® Linux Enterprise Server for High Performance Computing

Sep 05 | | The excellent scalability features of Linux, in addition to robust security and performance makes it an excellent choice for server systems, especially in the high performance computing area.

“Going Parallel - An Implementation Guide”

Sep 01 | | The paper outlines the basic steps and tools involved in the process of migrating a desktop application to a parallel environment.

Improving Performance and Manageability for Seismic Processing and Imaging Applications with Parallel Storage

Jun 05 | | As pressure increases on the upstream seismic processing community to deliver ever-higher levels of productivity and efficiency, a new generation of storage solutions will be required that allow the maximum utilisation of high-performance computing (HPC) Linux cluster resources, together with the minimum of management overhead.

View the White Paper Library

Multimedia

Video White Paper: Architecting a Better Network Storage Solution

BlueArc's Titan architecture represents an evolutionary step in file servers by creating a hardware-based file system that can scale bandwidth, IOPS, and overall data capacity well beyond conventional software-based devices. With its ability to virtualize a massive storage pool of up to four usable petabytes of tiered storage, Titan can scale with growing data requirements, offering a competitive advantage for businesses, researchers, or other enterprises seeking to better manage data growth while still ensuring optimal performance.

Podcast: Interview with Ben Bennett of ClearSpeed Technology

Today, HPC organizations are requiring substantially more floating point performance to solve real-world problems. In this podcast, Ben Bennett, ClearSpeed General Manager, discusses how acceleration technology can improve the overall performance of standard x86-based systems...