HPCwire: UltraSPARC T2 Finds a Home in HPC

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

HPCwire >> Blogs

Blog: From the Editor

June 11, 2008

UltraSPARC T2 Finds a Home in HPC

In all the excitement about the Roadrunner petaflop announcement this week, a bunch of other HPC news got pushed aside. One item that caught my eye was the announcement by the Canadian High Performance Computing Virtual Laboratory (HPCVL) that it had purchased a cluster made up of 78 Sun SPARC Enterprise T5140 servers, which is not a product you hear much about in the HPC space. In fact, it may be the only production system of its kind at an HPC facility.

The T5140 is a dual-socket server that uses the 8-core UltraSPARC T2 processor ("Niagara 2"). The T2 is the follow-on to the T1, which had only one floating point unit shared across its eight cores. When the T2 came along in 2007, Sun had remedied this by adding an FPU to each core, thereby making it suitable for technical computing.

The big deal about the T2 is that it offers lots of throughput in a very energy-efficient package. That's why the T2 servers are aimed mostly at enterprise users with scaled-out Web or data warehouse applications who want to consolidate resources. Since each processor core can handle 8 threads, the 78-node cluster the Canadians bought can juggle almost 10,000 simultaneously. Not bad for less than a 100 nodes.

The knock on the T2, at least for HPC, is a lack of raw performance. Each processor yields no more than 10 gigaflops or so on Linpack, mainly due to relatively slow clock speeds offered with the processor -- in the 0.9 to 1.4 GHz range. If an application can mostly run out of cache, Xeon or Opteron-based machines are going to outperform the UltraSPARC pretty handily.

Where T2 really shines is on highly multi-threaded codes that are limited by memory bandwidth, which is fairly common in real HPC codes. A good example is a PDE (partial differential equations) solver. In these cases, the T2 can make excellent use of the four on-chip memory controllers to speed access to RAM. Aggregate memory bandwidth per chip is advertised at 60+ GB/sec.

Last year, HPC researchers at Aachen University's Center for Computing and Communication (CCC) evaluated a pre-production system of a single-socket T2-based server against Woodcrest, Opteron, and UltraSPARC IV systems, using a number of benchmarks and application codes.

According to them, "The UltraSPARC T2 processor offers an amazing memory bandwidth, if multiple threads can be employed. And when parallelizing with OpenMP, the placement of threads and data is not critical, and also Solaris does a superb job in this respect already, whereas Linux on the Xeon and Opteron based system requires user attention." The complete evaluation by the Aachen group is available here.

If this kind of capability were encapsulated in an x86 processor, these would indeed be popular little chips today. The closest we'll get to an x86 version of the T2 will probably be a low-power, 8-core, Intel Nehalem processor sometime in 2009.

But by that time, Sun is expected to be offering its next-generation SPARC processor, called "Rock." Rock is a 16-core processor that will represent an entirely new architecture. The company says both thread performance and floating point performance will be better than the T2, and the processor will support new technologies like transactional memory and "scout threads." Sun originally wanted to deliver the processors this year, but is now targeting introduction for the second half of 2009.

Posted by Michael Feldman - June 11 @ 5:24PM

(Digg, Technorati, more)

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?

Michael Feldman

Michael Feldman is the editor of HPCwire.

More Michael Feldman

Feature Articles

The Week in Review

C-DAC announces plans for a petaflop system; IBM researchers are working on vertical integration techniques to extend Moore's Law another 15 years. We recap those stories and more in our weekly wrapup.
Read More...

Moscow State University Supercomputer Has Petaflop Aspirations

The Moscow State University supercomputer, Lomonosov, has been selected for a high-performance makeover, with the goal of tripling its processing power to achieve petaflop-level performance in 2010. T-Platforms, who developed and manufactured the supercomputer, is the odds-on favorite to lead the project.
Read More...

Intel Ups Performance Ante with Westmere Server Chips

Right on schedule, Intel has launched its Xeon 5600 processors, codenamed "Westmere EP." The 5600 represents the 32nm sequel to the Xeon 5500 (Nehalem EP) for dual-socket servers. Intel is touting better performance and energy efficiency, along with new security features, as the big selling points of the new Xeons.
Read More...

Read more HPCwire features...

Top Headlines

Australia Commissions Cray Supercomputer

Mar 19 | OfficialWire | New super to support intelligence work Down Under. Read more...

Intel Partners See 'Easy' Upgrade Path With Xeon 5600 Chips

Mar 18 | ChannelWeb | Westmere parts already showing up in HPC machines. Read more...

AMD: OEMs primed for Opteron 6100s

Mar 17 | The Register | But what about the tier ones? Read more...

Arrival of the Desktop Supercomputer

Mar 17 | Cadalyst Magazine | A new generation of workstations is changing the nature of technical computing. Read more...

Scheduling HPC In The Cloud

Mar 17 | Linux Magazine | Latest iteration of Sun Grid Engine able to tap into Cloud. 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.

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