Built for Speed,

By Ann Parker

December 16, 2005

While computer gamers are eagerly awaiting the next generation of platforms, the computer scientists of Lawrence Livermore's Graphics Architectures for Intelligence Applications (GAIA) project are tracking the rapidly changing technology, but for a different reason. A team, led by John Johnson of the Computation Directorate, is researching graphics processing units (GPUs)—the highly specialized, low-cost, rendering engines at the heart of the gaming industry—to determine how they might be programmed and used in applications other than virtual entertainment.

“Graphics processors are accelerating in performance much faster than other microprocessors,” says Sheila Vaidya, project leader for GAIA. “We have an opportunity to ride the wave of innovations driving the gaming industry.” These processors — traditionally designed for fast rendering of visual simulations, virtual reality, and computer gaming — could provide efficient solutions to some of the most challenging computing needs facing the intelligence and military communities. Real-time data-processing capabilities are needed for applications ranging from text and speech processing to image analysis for automated targeting and tracking.

Gaming the System

The GAIA team, including collaborators from Stanford University, the University of California at Berkeley and Davis, and Mississippi State University, is researching graphics processors used in the computer gaming and entertainment industries to determine how they might be used in knowledge-discovery applications of relevance to national security.

Why bother with this class of processors when plenty of central processing units (CPUs) exist to do the heavy-duty work in high-performance computing? Two words: speed and cost.

The ever-growing appetite in the three-dimensional (3D) interactive gaming community has led to the development and enhancement of GPUs at a rate faster than the performance of conventional microprocessors predicted by Moore's Law. This acceleration in improved performance will likely continue as long as the demand exists and integrated-circuit technologies continue to scale.

During the past 2 years, the GAIA team has implemented many algorithms on current-generation CPUs and GPUs to compare their performance. The benchmarks that followed showed amazing performance gains of one to two orders of magnitude on GPUs for a variety of applications, such as georegistration, hyperspectral imaging, speech recognition, image processing, bioinformatics, and seismic exploration.
 
GPUs have a number of features that make them attractive for both image- and data-processing applications. For example, they are designed to exploit the highly parallel nature of graphics-rendering algorithms, and they efficiently use the hundreds of processing units available on-chip for parallel computing. Thus, one operation can be simultaneously performed on multiple data sets in an architecture known as single-instruction, multiple data (SIMD), providing extremely high-performance arithmetic capabilities for specific classes of applications. Current high-end GPU chips can handle up to 24 pipelines of data per chip and perform hundreds of billions of operations per second.
 
Today's commercial GPUs are relatively inexpensive as well. “National retailers charge a few hundred dollars for one, compared to the thousands of dollars or more that a custom-built coprocessor might cost,” says Johnson.

The performance of these GPUs is impressive when compared with that of even the newest CPUs. “A modern CPU performs about 25 billion floating-point operations per second,” says Johnson. “Whereas a leading-edge GPU, such as the NVIDIA GeForce 7800 GTX video card or the upcoming successor to the ATI Radeon X850, performs six times faster at half the cost of a CPU.” These GPUs are optimized for calculating the floating-point arithmetic associated with 3D graphics and for performing large numbers of operations simultaneously.

GPUs also feature a high on-chip memory bandwidth, that is, a large data-carrying capacity, and have begun to support more advanced instructions used in general-purpose computing. When combined with conventional CPUs and some artful programming, these devices could be used for a variety of high-throughput applications.

“GPUs work well on problems that can be broken down into many small, independent tasks,” explains GAIA team member Dave Bremer. Each task in the problem is matched with a pixel in an output image. A short program is loaded into the GPU, which is executed once for every pixel drawn, and the results from each execution are stored in an image. As the image is being drawn, many tasks are being executed simultaneously through the GPU's numerous pipelines. Finally, the results of the problem are copied back to an adjacent CPU.

However, general-purpose programming on GPUs still poses significant challenges. Because the tasks performed on a GPU occur in an order that is not controlled by a programmer, no one task can depend on the results of a previous one, and tasks cannot write to the same memory. Consequently, image convolution operations work extremely well (100 times faster) because output pixels are computed independently, but computing a global sum becomes very complex because there is no shared memory. “Data must be copied in and out of the GPU over a relatively slow transmission path,” says GAIA team member Jeremy Meredith. “As a result, memory-intensive computations that require arbitrary access to large amounts of memory off-chip are not well suited to the GPU architecture.”

Today's GPUs are power hungry. But designers, faced with the growing demand for mobile computing, are rapidly evolving chip architectures to develop low-power versions that will approach the performance of high-end workstations.

What's in the Pipeline

“GPUs are beginning to more closely resemble CPUs with every evolution,” notes Johnson. “The drawbacks for general-purpose programming are being tackled by the industry, one by one.” Next-generation CPU architectures are adopting many features from GPUs. “Emerging architectural designs such as those found in Stanford's Merrimac and the IBM-Toshiba-Sony Cell processor look similar to the architecture of GPUs,” says Johnson. “These designs could be the next-generation technology for real-time, data-processing applications. Our work with GPUs will help us evaluate and deploy the emerging devices.”

The Cell processor, which is a crossover GPU-CPU chip, is scheduled to hit the gaming market soon. But the Cell might also prove to be useful in defense and security computing environments. The scientists of GAIA — just like the gamers — are eager to test and scale its limits.

For further information contact John Johnson (925) 424-4092 (jjohnson@llnl.gov) or Sheila Vaidya (925) 423-5428 (vaidya1@llnl.gov).

Credit must be given to the University of California, Lawrence Livermore National Laboratory, and the Department of Energy under whose auspices the work was performed, when this information or a reproduction of it is used.

Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industy updates delivered to you every week!

ExxonMobil, NCSA, Cray Scale Reservoir Simulation to 700,000+ Processors

February 17, 2017

In a scaling breakthrough for oil and gas discovery, ExxonMobil geoscientists report they have harnessed the power of 717,000 processors – the equivalent of 22,000 32-processor computers – to run complex oil and gas reservoir simulation models. Read more…

By Doug Black

TSUBAME3.0 Points to Future HPE Pascal-NVLink-OPA Server

February 17, 2017

Since our initial coverage of the TSUBAME3.0 supercomputer yesterday, more details have come to light on this innovative project. Of particular interest is a new board design for NVLink-equipped Pascal P100 GPUs that will create another entrant to the space currently occupied by Nvidia's DGX-1 system, IBM's "Minsky" platform and the Supermicro SuperServer (1028GQ-TXR). Read more…

By Tiffany Trader

Tokyo Tech’s TSUBAME3.0 Will Be First HPE-SGI Super

February 16, 2017

In a press event Friday afternoon local time in Japan, Tokyo Institute of Technology (Tokyo Tech) announced its plans for the TSUBAME3.0 supercomputer, which will be Japan’s “fastest AI supercomputer,” Read more…

By Tiffany Trader

Drug Developers Use Google Cloud HPC in the Fight Against ALS

February 16, 2017

Within the haystack of a lethal disease such as ALS (amyotrophic lateral sclerosis / Lou Gehrig’s Disease) there exists, somewhere, the needle that will pierce this therapy-resistant affliction. Read more…

By Doug Black

HPE Extreme Performance Solutions

Object Storage is the Ideal Storage Method for CME Companies

The communications, media, and entertainment (CME) sector is experiencing a massive paradigm shift driven by rising data volumes and the demand for high-performance data analytics. Read more…

Weekly Twitter Roundup (Feb. 16, 2017)

February 16, 2017

Here at HPCwire, we aim to keep the HPC community apprised of the most relevant and interesting news items that get tweeted throughout the week. Read more…

By Thomas Ayres

Alexander Named Dep. Dir. of Brookhaven Computational Initiative

February 15, 2017

Francis Alexander, a physicist with extensive management and leadership experience in computational science research, has been named Deputy Director of the Computational Science Initiative at the U.S. Read more…

Here’s What a Neural Net Looks Like On the Inside

February 15, 2017

Ever wonder what the inside of a machine learning model looks like? Today Graphcore released fascinating images that show how the computational graph concept maps to a new graph processor and graph programming framework it’s creating. Read more…

By Alex Woodie

Azure Edges AWS in Linpack Benchmark Study

February 15, 2017

The “when will clouds be ready for HPC” question has ebbed and flowed for years. Read more…

By John Russell

TSUBAME3.0 Points to Future HPE Pascal-NVLink-OPA Server

February 17, 2017

Since our initial coverage of the TSUBAME3.0 supercomputer yesterday, more details have come to light on this innovative project. Of particular interest is a new board design for NVLink-equipped Pascal P100 GPUs that will create another entrant to the space currently occupied by Nvidia's DGX-1 system, IBM's "Minsky" platform and the Supermicro SuperServer (1028GQ-TXR). Read more…

By Tiffany Trader

Tokyo Tech’s TSUBAME3.0 Will Be First HPE-SGI Super

February 16, 2017

In a press event Friday afternoon local time in Japan, Tokyo Institute of Technology (Tokyo Tech) announced its plans for the TSUBAME3.0 supercomputer, which will be Japan’s “fastest AI supercomputer,” Read more…

By Tiffany Trader

Drug Developers Use Google Cloud HPC in the Fight Against ALS

February 16, 2017

Within the haystack of a lethal disease such as ALS (amyotrophic lateral sclerosis / Lou Gehrig’s Disease) there exists, somewhere, the needle that will pierce this therapy-resistant affliction. Read more…

By Doug Black

Azure Edges AWS in Linpack Benchmark Study

February 15, 2017

The “when will clouds be ready for HPC” question has ebbed and flowed for years. Read more…

By John Russell

Is Liquid Cooling Ready to Go Mainstream?

February 13, 2017

Lost in the frenzy of SC16 was a substantial rise in the number of vendors showing server oriented liquid cooling technologies. Three decades ago liquid cooling was pretty much the exclusive realm of the Cray-2 and IBM mainframe class products. That’s changing. We are now seeing an emergence of x86 class server products with exotic plumbing technology ranging from Direct-to-Chip to servers and storage completely immersed in a dielectric fluid. Read more…

By Steve Campbell

Cray Posts Best-Ever Quarter, Visibility Still Limited

February 10, 2017

On its Wednesday earnings call, Cray announced the largest revenue quarter in the company’s history and the second-highest revenue year. Read more…

By Tiffany Trader

HPC Cloud Startup Launches ‘App Store’ for HPC Workflows

February 9, 2017

“Civilization advances by extending the number of important operations which we can perform without thinking about them,” Read more…

By Tiffany Trader

Intel and Trump Announce $7B for Fab 42 Targeting 7nm

February 8, 2017

In what may be an attempt by President Trump to reset his turbulent relationship with the high tech industry, he and Intel CEO Brian Krzanich today announced plans to invest more than $7 billion to complete Fab 42. Read more…

By John Russell

For IBM/OpenPOWER: Success in 2017 = (Volume) Sales

January 11, 2017

To a large degree IBM and the OpenPOWER Foundation have done what they said they would – assembling a substantial and growing ecosystem and bringing Power-based products to market, all in about three years. Read more…

By John Russell

US, China Vie for Supercomputing Supremacy

November 14, 2016

The 48th edition of the TOP500 list is fresh off the presses and while there is no new number one system, as previously teased by China, there are a number of notable entrants from the US and around the world and significant trends to report on. Read more…

By Tiffany Trader

Lighting up Aurora: Behind the Scenes at the Creation of the DOE’s Upcoming 200 Petaflops Supercomputer

December 1, 2016

In April 2015, U.S. Department of Energy Undersecretary Franklin Orr announced that Intel would be the prime contractor for Aurora: Read more…

By Jan Rowell

D-Wave SC16 Update: What’s Bo Ewald Saying These Days

November 18, 2016

Tucked in a back section of the SC16 exhibit hall, quantum computing pioneer D-Wave has been talking up its new 2000-qubit processor announced in September. Forget for a moment the criticism sometimes aimed at D-Wave. This small Canadian company has sold several machines including, for example, ones to Lockheed and NASA, and has worked with Google on mapping machine learning problems to quantum computing. In July Los Alamos National Laboratory took possession of a 1000-quibit D-Wave 2X system that LANL ordered a year ago around the time of SC15. Read more…

By John Russell

Enlisting Deep Learning in the War on Cancer

December 7, 2016

Sometime in Q2 2017 the first ‘results’ of the Joint Design of Advanced Computing Solutions for Cancer (JDACS4C) will become publicly available according to Rick Stevens. He leads one of three JDACS4C pilot projects pressing deep learning (DL) into service in the War on Cancer. Read more…

By John Russell

HPC Startup Advances Auto-Parallelization’s Promise

January 23, 2017

The shift from single core to multicore hardware has made finding parallelism in codes more important than ever, but that hasn’t made the task of parallel programming any easier. Read more…

By Tiffany Trader

IBM Wants to be “Red Hat” of Deep Learning

January 26, 2017

IBM today announced the addition of TensorFlow and Chainer deep learning frameworks to its PowerAI suite of deep learning tools, which already includes popular offerings such as Caffe, Theano, and Torch. Read more…

By John Russell

CPU Benchmarking: Haswell Versus POWER8

June 2, 2015

With OpenPOWER activity ramping up and IBM’s prominent role in the upcoming DOE machines Summit and Sierra, it’s a good time to look at how the IBM POWER CPU stacks up against the x86 Xeon Haswell CPU from Intel. Read more…

By Tiffany Trader

Leading Solution Providers

Nvidia Sees Bright Future for AI Supercomputing

November 23, 2016

Graphics chipmaker Nvidia made a strong showing at SC16 in Salt Lake City last week. Read more…

By Tiffany Trader

BioTeam’s Berman Charts 2017 HPC Trends in Life Sciences

January 4, 2017

Twenty years ago high performance computing was nearly absent from life sciences. Today it’s used throughout life sciences and biomedical research. Genomics and the data deluge from modern lab instruments are the main drivers, but so is the longer-term desire to perform predictive simulation in support of Precision Medicine (PM). There’s even a specialized life sciences supercomputer, ‘Anton’ from D.E. Shaw Research, and the Pittsburgh Supercomputing Center is standing up its second Anton 2 and actively soliciting project proposals. There’s a lot going on. Read more…

By John Russell

Container App ‘Singularity’ Eases Scientific Computing

October 20, 2016

HPC container platform Singularity is just six months out from its 1.0 release but already is making inroads across the HPC research landscape. It's in use at Lawrence Berkeley National Laboratory (LBNL), where Singularity founder Gregory Kurtzer has worked in the High Performance Computing Services (HPCS) group for 16 years. Read more…

By Tiffany Trader

Dell Knights Landing Machine Sets New STAC Records

November 2, 2016

The Securities Technology Analysis Center, commonly known as STAC, has released a new report characterizing the performance of the Knight Landing-based Dell PowerEdge C6320p server on the STAC-A2 benchmarking suite, widely used by the financial services industry to test and evaluate computing platforms. The Dell machine has set new records for both the baseline Greeks benchmark and the large Greeks benchmark. Read more…

By Tiffany Trader

IDG to Be Bought by Chinese Investors; IDC to Spin Out HPC Group

January 19, 2017

US-based publishing and investment firm International Data Group, Inc. (IDG) will be acquired by a pair of Chinese investors, China Oceanwide Holdings Group Co., Ltd. Read more…

By Tiffany Trader

What Knights Landing Is Not

June 18, 2016

As we get ready to launch the newest member of the Intel Xeon Phi family, code named Knights Landing, it is natural that there be some questions and potentially some confusion. Read more…

By James Reinders, Intel

KNUPATH Hermosa-based Commercial Boards Expected in Q1 2017

December 15, 2016

Last June tech start-up KnuEdge emerged from stealth mode to begin spreading the word about its new processor and fabric technology that’s been roughly a decade in the making. Read more…

By John Russell

Intel and Trump Announce $7B for Fab 42 Targeting 7nm

February 8, 2017

In what may be an attempt by President Trump to reset his turbulent relationship with the high tech industry, he and Intel CEO Brian Krzanich today announced plans to invest more than $7 billion to complete Fab 42. Read more…

By John Russell

  • arrow
  • Click Here for More Headlines
  • arrow
Share This