HPCwire: China Joins Petaflop Club

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

HPCwire >> Blogs

Blog: From the Editor

October 29, 2009

China Joins Petaflop Club

China has apparently become the third country to build a petaflop supercomputer. Xinhua news agency reported on Thursday that the country has unveiled "Tianhe," a 1.206 peak petaflop machine powered by a combination of 6,144 Intel CPUs and 5,120 AMD GPUs. Amazingly, the price tag was a mere $88.24 million. The system is installed at the National University of Defense Technology (NUDT) in Changsha, capital of central China's Hunan Province.

In the TOP500 sense, Tianhe would not be considered a true petaflop system. According to online reports, the machine achieves only(!) 563.1 teraflops with Linpack. If that number holds up, it would almost certainly earn Tianhe a spot in the top 10 of the upcoming TOP500 list. Today there are only three systems that break the 500 teraflop barrier on Linpack: Roadrunner at Los Alamos National Laboratory, Jaguar at Oak Ridge National Laboratory, and JUGENE at Jülich Supercomputing Center. China's top system on the current list is Dawning's "Magic Cube" supercomputer, located at the Shanghai Supercomputer Center. With a Linpack rating of 180.6 teraflops, the Dawning machine sits at number 15.

At some future date, NUDT is slated to add "hundreds or thousands of China-made CPUs to the machine, and improve its Linpack performance to over 800 teraflops," according to Zhou Xingming, an academician in the Chinese Academy of Sciences and a professor at NUDT.

The Xinhua news article, as well as other early reports from Chinese sources, don't provide much detail about the system's architecture. Specifically, no information was offered about the kind of Intel CPU and AMD GPU parts used, nor about the Chinese-made CPUs to be plugged in later on. At press time, NUDT could not be reached for further clarification about Tianhe's make-up, and AMD declined to offer any additional details.

If I had to speculate, I would guess that the Intel chips are Nehalem EPs and the AMD parts are FireStream 9270s. Presumably most of the FLOPS come from the GPUs. In fact, 5,000 9270s would represent 1.2 double precision petaflops all by themselves. The future Chinese CPUs are likely to be of the Godson-3 variety, which are expected to debut in 2010. Note that the Godson-3 is a MIPS architecture, but has the capability to emulate x86 instructions as well.

In addition to the sketchy details on the architecture, no mention was made of the application set the machine will be targeting. NUDT is jointly run by the Ministry of National Defense and the Ministry of Education, which gives you some idea of its areas of interest. According the university's Web site, the institution is devoted to basic sciences, engineering, military science, management, economics, philosophy, literature, education, law, and history.

Impressive as this all sounds, Tianhe's rather low Linpack efficiency (Rmax/Rpeak) may limit its applicability somewhat. Linpack usually represents a nominal high-water mark for what kind of performance you're likely to get from math-intensive applications. The NUDT machine didn't even manage to reach the 50 percent mark in efficiency -- just 563 out of a possible 1206 teraflops. Most supers have a Linpack efficiency north of 75 percent, even just for vanilla GigE clusters. The new Earth Simulator in Japan boasts a 93.4 percent figure.

Undoubtedly, the problem is related to extracting Linpack FLOPS from the GPUs. Although one would think these general purpose graphics processors would excel at this type of vector math, optimal Linpack performance is also dependent on a generous cache. Modern CPUs have plenty of it, but GPUs contain only limited internal caches. That means the graphics chip would have to access the relatively slower on-board GDDR memory to refresh its data, or worse yet, go across the PCIe bus to get some more data from CPU memory. NVIDIA's upcoming Fermi processor will be the first GPU with a true cache hierarchy (not to mention much better double precision performance), so I imagine Linpack results on this architecture should be a good deal more impressive.

In the meantime, Tianhe will represent an interesting test case for a CPU-GPU hybrid supercomputer, an architecture which is likely to become more commonplace over the next few years. It also signals China's intention to become a bigger player in the supercomputing arena. Given the country's huge cash reserves and the government's willingness to invest in high-tech, there's not much that can stop it.

Posted by Michael Feldman - October 29 @ 6:41PM

(Digg, Technorati, more)

Discussion

There are 1 discussion items posted.

563 out of 1206
Submitted by vvolkov on 10/30/2009 - 1:58AM

Small fraction of peak may not be due to the lack of caches (in fact, GPUs have enough of other on-chip memory, such as ~2MB register files), but due to an inefficient implementation of matrix multiply.

Here is a recent story with ATI's matrix multiply, albeit in single precision. Original ATI's implementation was running at 45% of peak on HD4870 and was claimed optimal. However, somebody has recently claimed achieving 82% of peak by using larger register blocking. It turned out (according to an ATI/AMD engineer) that the original code was optimized for an earlier GPU (HD3870) and was not updated since then.

I suggest same applies to their double precision matrix multiply. Improve it along the same lines and get a petaflop.

Post #1

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

TACC's Ranger supercomputer celebrates its second year of enabling important research; Microsoft partners with NSF to bring cloud services to researchers; and NSF submits its fiscal year 2011 budget request. We recap those stories and more in our weekly wrapup.
Read More...

NASA Looks to Move Science Apps Into the Cloud

It seems only natural that the US space agency would be casting its eyes toward the clouds. Sure enough, NASA is now looking to cloud computing to optimize the operation of the agency's IT infrastructure for some of its science codes. Like many commercial businesses and government organizations, NASA is being asked to do more computing with fewer datacenter resources.
Read More...

Thoughts, Observations, Beliefs & Opinions About the NSF Supercomputer Centers

There is no such thing as an NSF (Supercomputer) Center and there never has been. There should be. What there are, in the words of Ed Hayes, then comptroller of NSF, are "NSF ASSISTED Supercomputer Centers." This is a double edged sword.
Read More...

Read more HPCwire features...

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