Convey Computer President’s Letter

By Nicole Hemsoth

January 19, 2010

The new year is a time for reflection, especially after the turbulent economic environment that marked 2009. In that spirit, HPCwire invited the HPC vendor community to post a new year’s message to our readers. First in the series is Bruce Toal, CEO, president and co-founder for Convey Computer.

Bruce Toal, CEO, President, Co-Founder Convey ComputerDear HPCwire Readers, Customers, and Colleagues,

For a new computer company, 2009 was a year of proving that technical innovation could thrive in the crushingly competitive high-performance computing marketplace. It was our year to acquire more investment capital, more customers, and release the production units of our hybrid-core computer into the marketplace.

Happily, 2009 was a remarkable year for Convey Computer Corporation. As pioneers of hybrid-core computing – a technology designed to help high-performance computers run faster while significantly reducing the energy they use – we closed $24.15 million in Series B financing in July. The round of funding was led by new investor, Braemar Energy Ventures. Series A investors CenterPoint Ventures, Intel Capital, InterWest Partners, Rho Ventures, and Xilinx® also participated. Convey was founded in 2006 and introduced its product, the Convey HC-1™, in November 2008. To date, Convey has received $39.25 million in venture funding.

Our investors were impressed that the Convey team met its milestones, including recently shipping beta and production units to customers who are using the systems in diverse applications such as bioinformatics, data mining, government, and oil and gas. In 2009, Convey announced two new board members with extensive technology experience: Jiong Ma, Ph.D., of Braemar Energy Ventures, and Joshua Ruch of Rho Ventures.

The new round of investment capital meant that we could begin to expand our employee infrastructure. We now have a talented and knowledgeable sales force with proven expertise in the high-performance computing. At the same time, we are building our customer support organization and other company functions to deal with scaling production. Convey also expanded internationally in 2009 when we signed a master reseller agreement with HMK Supercomputing GmbH, whichwill represent Convey Computer in Germany, Switzerland, and Austria.

2010: The Focus Sharpens on Green HPC

“Hybrid-core” is a new market space identified by Convey. A hybrid-core computer improves application performance by combining an x86 core with hardware that implements application-specific instructions. Hybrid-core computing was developed to address what we saw as a perfect storm of technology trends, including:

  1. The need to continue extracting application performance from hardware rather than transferring the burden to programmers
  2. The flattening — or even the decline — in processor clock rates
  3. The need to reduce the cost and the use of energy in our data centers

The energy issue is paramount. Convey’s hybrid-core technology helps customers reduce energy costs associated with high-performance computing because we use FPGAs to power our compute engines. To illustrate: a modern data center today can cost up to $200 million to build, with power and cooling costs making up half of the annual operating costs. Every watt required to power a server nominally requires another watt to cool it. On key HPC workloads, Convey’s HC-1 reduces the number of servers required dramatically. One rack of Convey servers can replace multiple racks of traditional application servers, which means that customers can realize dramatically lower power and cooling costs. One of our customers, for example, realized a 91% reduction in their power and cooling costs. And they’re pretty happy about that!

Reducing power without sacrificing performance is one of the most important issues facing the HPC community over the coming years. The reason Convey will continue to focus on “Green HPC” is simple: data center economics have changed. Energy costs are consuming half the IT operating budget and the cost of floor space is limiting. As we’ve heard from Dr. John Shalf, head of Berkeley Lab’s Science-Driven Systems Architecture team, “Energy efficiency has become a first-order design constraint for future systems. We really don’t see the current path of scaling up conventional hardware as sustainable either in terms of the initial hardware cost or the price of powering such systems over its lifetime.

2010: Where’s the Software?

The same issues that have been driving high-performance computing for the last couple of decades – more performance/less price – will continue into 2010 and beyond. In our view, the performance issue is one part hardware and two parts software.

Our motto at Convey is that the system which is easier to program wins. Convey is an application-specific, low power node. What continues to challenge HPC is that the cost of a programmer for one year is MORE than the cost of acquiring a teraflop (peak performance) system. This is one reason we engineered the Convey system with the hardware and software architecture as an extension to the x86 ecosystem. When presented with a familiar environment, programmers benefit from 100 percent productivity and portability.

Many applications have code that was written long ago – in some case 20 years ago! Many of these applications still have “serial math” as their underpinning. And, before we forget, the universe of programmers capable of writing parallel programs does not seem to have grown any over the years. Now, HPC applications are moving into a different application space called data-intensive computing. The computer centers at Google and Microsoft are substantially larger than what was once thought of as a “classic” HPC center. For example, Microsoft’s Chicago datacenter is 24,000 square meters, houses some one million servers, and is powered by two 300MW substations!

The industry needs to focus more on software productivity, just as we are doing at Convey. Admittedly, it takes time and money to redefine. Meanwhile, extracting performance from our portfolio of HPC applications is easiest for the programmer if the compiler does the work or, at minimum, provides the hints that point the programmer in the right direction. Plus, having a full suite of development tools available to help in the debug and optimization phases of development helps a lot. We believe this is another area where our cache coherent, virtual memory approach really shines – in being able to step through all phases of the program, whether executing on the x86 or on the Convey coprocessor in a single process space.

2010: New Architectures Bring New Opportunities for Discovery in HPC

As the future continues to bring innovation to the HPC marketplace, we will also continue to innovate with hybrid-core technologies.From our perspective, computer architectures will continue to evolve with a greater emphasis on application flexibility.

Fundamental to our mission is to design and develop personalities with instructions that can replace thousands of iterations of standard x86 instruction for a particular application space and with full ANSI standard FORTRAN, C, and C++compiler support.Our technology stimulates the imaginative talents of computer architects and algorithm experts. It is no longer an exercise in using a standard instruction set for optimal performance, but a creative challenge to define a set of instructions that are optimized for a set of applications. The innovation continues as computing experts can “teach” a compiler how to recognize certain program constructs and then invoke these instructions.

We saw this first-hand as we worked in 2009 to develop the new Financial Analytics Personality (FAP). One of the major problems facing developers of financial applications is that the instruction set of commodity processors doesn’t map well to many financial algorithms. Even a simple intrinsic function can take hundreds of instructions and, in many of these applications, that intrinsic function must be executed millions of times.

The Convey solution hard-wires these functions into the HC-1system hardware, which can be many times more efficient than a sequence of commodity instructions. The ANSI standard Convey compilers automatically recognize constructs in the application that can take advantage of the hardware speed-ups. In effect, the developer doesn’t need to know the intricacies of the hardware, yet they reap the performance benefits.

Application-specific instructions are not practical to implement within a microprocessor unless the market that can take advantage of them is sizable, which means large enough to cover the three or five year development costs. Potential application-specific instructions must then compete for processor die space with other uses for transistors (more cache, more cores, etc.). Using a reconfigurable fabric (FPGAs) reduces the design cycle to months instead of years and avoids the huge costs to produce a custom device. And reconfigurable fabric can be used for many application areas, each optimized for that specific mission without competing for device real estate against other uses or areas.

This is one reason why we believe that FPGAs – with the technology’s inherent application flexibility and energy efficiency – represent the best solution for delivering more raw performance, nimble application development, and energy-efficient computing.

In closing, here’s wishing all of us a very exciting 2010.

Best Regards,

Bruce Toal
CEO, President, Co-Founder
Convey Computer Corporation
Convey Computer

Subscribe to HPCwire's Weekly Update!

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

MLPerf Inference 4.0 Results Showcase GenAI; Nvidia Still Dominates

March 28, 2024

There were no startling surprises in the latest MLPerf Inference benchmark (4.0) results released yesterday. Two new workloads — Llama 2 and Stable Diffusion XL — were added to the benchmark suite as MLPerf continues Read more…

Q&A with Nvidia’s Chief of DGX Systems on the DGX-GB200 Rack-scale System

March 27, 2024

Pictures of Nvidia's new flagship mega-server, the DGX GB200, on the GTC show floor got favorable reactions on social media for the sheer amount of computing power it brings to artificial intelligence.  Nvidia's DGX Read more…

Call for Participation in Workshop on Potential NSF CISE Quantum Initiative

March 26, 2024

Editor’s Note: Next month there will be a workshop to discuss what a quantum initiative led by NSF’s Computer, Information Science and Engineering (CISE) directorate could entail. The details are posted below in a Ca Read more…

Waseda U. Researchers Reports New Quantum Algorithm for Speeding Optimization

March 25, 2024

Optimization problems cover a wide range of applications and are often cited as good candidates for quantum computing. However, the execution time for constrained combinatorial optimization applications on quantum device Read more…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at the network layer threatens to make bigger and brawnier pro Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HBM3E memory as well as the the ability to train 1 trillion pa Read more…

MLPerf Inference 4.0 Results Showcase GenAI; Nvidia Still Dominates

March 28, 2024

There were no startling surprises in the latest MLPerf Inference benchmark (4.0) results released yesterday. Two new workloads — Llama 2 and Stable Diffusion Read more…

Q&A with Nvidia’s Chief of DGX Systems on the DGX-GB200 Rack-scale System

March 27, 2024

Pictures of Nvidia's new flagship mega-server, the DGX GB200, on the GTC show floor got favorable reactions on social media for the sheer amount of computing po Read more…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HB Read more…

Nvidia Looks to Accelerate GenAI Adoption with NIM

March 19, 2024

Today at the GPU Technology Conference, Nvidia launched a new offering aimed at helping customers quickly deploy their generative AI applications in a secure, s Read more…

The Generative AI Future Is Now, Nvidia’s Huang Says

March 19, 2024

We are in the early days of a transformative shift in how business gets done thanks to the advent of generative AI, according to Nvidia CEO and cofounder Jensen Read more…

Nvidia’s New Blackwell GPU Can Train AI Models with Trillions of Parameters

March 18, 2024

Nvidia's latest and fastest GPU, codenamed Blackwell, is here and will underpin the company's AI plans this year. The chip offers performance improvements from Read more…

Nvidia Showcases Quantum Cloud, Expanding Quantum Portfolio at GTC24

March 18, 2024

Nvidia’s barrage of quantum news at GTC24 this week includes new products, signature collaborations, and a new Nvidia Quantum Cloud for quantum developers. Wh Read more…

Alibaba Shuts Down its Quantum Computing Effort

November 30, 2023

In case you missed it, China’s e-commerce giant Alibaba has shut down its quantum computing research effort. It’s not entirely clear what drove the change. Read more…

Nvidia H100: Are 550,000 GPUs Enough for This Year?

August 17, 2023

The GPU Squeeze continues to place a premium on Nvidia H100 GPUs. In a recent Financial Times article, Nvidia reports that it expects to ship 550,000 of its lat Read more…

Shutterstock 1285747942

AMD’s Horsepower-packed MI300X GPU Beats Nvidia’s Upcoming H200

December 7, 2023

AMD and Nvidia are locked in an AI performance battle – much like the gaming GPU performance clash the companies have waged for decades. AMD has claimed it Read more…

DoD Takes a Long View of Quantum Computing

December 19, 2023

Given the large sums tied to expensive weapon systems – think $100-million-plus per F-35 fighter – it’s easy to forget the U.S. Department of Defense is a Read more…

Synopsys Eats Ansys: Does HPC Get Indigestion?

February 8, 2024

Recently, it was announced that Synopsys is buying HPC tool developer Ansys. Started in Pittsburgh, Pa., in 1970 as Swanson Analysis Systems, Inc. (SASI) by John Swanson (and eventually renamed), Ansys serves the CAE (Computer Aided Engineering)/multiphysics engineering simulation market. Read more…

Choosing the Right GPU for LLM Inference and Training

December 11, 2023

Accelerating the training and inference processes of deep learning models is crucial for unleashing their true potential and NVIDIA GPUs have emerged as a game- Read more…

Intel’s Server and PC Chip Development Will Blur After 2025

January 15, 2024

Intel's dealing with much more than chip rivals breathing down its neck; it is simultaneously integrating a bevy of new technologies such as chiplets, artificia Read more…

Baidu Exits Quantum, Closely Following Alibaba’s Earlier Move

January 5, 2024

Reuters reported this week that Baidu, China’s giant e-commerce and services provider, is exiting the quantum computing development arena. Reuters reported � Read more…

Leading Solution Providers

Contributors

Comparing NVIDIA A100 and NVIDIA L40S: Which GPU is Ideal for AI and Graphics-Intensive Workloads?

October 30, 2023

With long lead times for the NVIDIA H100 and A100 GPUs, many organizations are looking at the new NVIDIA L40S GPU, which it’s a new GPU optimized for AI and g Read more…

Shutterstock 1179408610

Google Addresses the Mysteries of Its Hypercomputer 

December 28, 2023

When Google launched its Hypercomputer earlier this month (December 2023), the first reaction was, "Say what?" It turns out that the Hypercomputer is Google's t Read more…

AMD MI3000A

How AMD May Get Across the CUDA Moat

October 5, 2023

When discussing GenAI, the term "GPU" almost always enters the conversation and the topic often moves toward performance and access. Interestingly, the word "GPU" is assumed to mean "Nvidia" products. (As an aside, the popular Nvidia hardware used in GenAI are not technically... Read more…

Shutterstock 1606064203

Meta’s Zuckerberg Puts Its AI Future in the Hands of 600,000 GPUs

January 25, 2024

In under two minutes, Meta's CEO, Mark Zuckerberg, laid out the company's AI plans, which included a plan to build an artificial intelligence system with the eq Read more…

Google Introduces ‘Hypercomputer’ to Its AI Infrastructure

December 11, 2023

Google ran out of monikers to describe its new AI system released on December 7. Supercomputer perhaps wasn't an apt description, so it settled on Hypercomputer Read more…

China Is All In on a RISC-V Future

January 8, 2024

The state of RISC-V in China was discussed in a recent report released by the Jamestown Foundation, a Washington, D.C.-based think tank. The report, entitled "E Read more…

Intel Won’t Have a Xeon Max Chip with New Emerald Rapids CPU

December 14, 2023

As expected, Intel officially announced its 5th generation Xeon server chips codenamed Emerald Rapids at an event in New York City, where the focus was really o Read more…

IBM Quantum Summit: Two New QPUs, Upgraded Qiskit, 10-year Roadmap and More

December 4, 2023

IBM kicks off its annual Quantum Summit today and will announce a broad range of advances including its much-anticipated 1121-qubit Condor QPU, a smaller 133-qu Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire