The Microsoft of FPGAs

By Michael Feldman

October 27, 2006

HPC and modern computing in general has a seemingly insatiable demand for more performance, better efficiency and scalability. Now with the expansion of computing to practically every commercial and non-commercial endeavor, an additional requirement is to apply these attributes to a much wider range of applications.

“So there's a need there for more types of processing,” says Jeff Jussel, vice president of marketing and general manager of the Americas for Celoxica. “There's a number of ways the industry is addressing that — with massively parallel processing (MPP) and with all sorts of different types of co-processors. At Celoxica, we believe that the FPGA represents a huge opportunity for co-processing, because it can deliver the massive parallelization, with the advantages of custom hardware, but in a way that is programmable.”

But there are three things that you need in order for FPGAs to really take off:

  1. Capable hardware: You've got to have the FPGAs that are powerful enough to handle the load, and do something interesting. Today that's not a problem. For example, even the standard technology Xilinx Virtex 4 (90nm) devices have the capability to run workloads from the financial services, oil and gas, and life science spaces with much higher performance than conventional scalar processors.
  2. High performance interconnects: You can't just use a PCI bus to communicate with the processor to provide the level of bandwidth and latency required to offload anything very interesting. But with PCI Express and HyperTransport, that level of interconnect performance is now available.
  3. Programmability: You can't expect software developers on Wall Street to understand RTL (register transfer level) technology to program the hardware. Familiar software tools and programming models need to be available. This is the biggest challenge for FPGAs today.

“That's where the tools that Celoxica provides comes in,” says Jussel. “That's our mission in life — to make that FPGA programming transparent. And in doing so, enable FPGA use for the high performance computing market.”

As part of this strategy, this week Celoxica announced a new off-the-shelf hardware and software compiler design bundle for high performance computing using HyperTransport (HTX) slots. The HTX bundle combines an intellectual property (IP) core for HTX connectivity, an FPGA-based HTX acceleration card and a software programming environment. The solution is designed to allow users to accelerate applications in Opteron-based computing systems with FPGA co-processing and HyperTransport technology. The bundle provides compilers that map C code onto FPGA hardware, a run-time OS (RTOS) for FPGA computing, and FPGA hardware that plugs into a host server system.

The hardware consists of the RCHTX acceleration card, which includes two Xilinx Virtex 4 FPGAs devices (in the future it will support more advanced Virtex 5 devices), 24 MB of QDR SRAM, and a range of I/O. The main co-processor FPGA is a 16 million gate device that is meant to run the user algorithms. The second FPGA is configured as a bridge, containing an HTX IP core developed by Celoxica. The bridge FPGA and IP provide the HyperTransport interconnect between the FPGA co-processor and the host processor system and memory space.

The software component consists of the DK Design Suite, which includes a C compiler for programming the FPGA co-processor, a board support package (BSP) and data communications drivers for the RCHTX card, a basic floating point library (single and double precision) and the software API which provides the interfaces.

The idea is not for the user to port their whole application to the FPGAs, just the compute-intensive algorithms that represent the workload bottlenecks. For example FFT calculations, a Black Scholes algorithm or wave migration calculations can be offloaded to the FPGA to take advantage of the parallel hardware resources.

The user replaces the algorithm loop in the original FORTRAN or C source with a Celoxica API call, which calls the C code that will be compiled into the FPGA. Jussel says the original algorithm needs to be “tweaked” somewhat to insert parallelism, but they've tried to simplify this as much as possible. The FPGA C compiler brings in the appropriate run-time pieces to make it work in its new hardware environment. At execution time, the data communication between the host processor and the FPGA is done across the HyperTransport connection, but this is transparent to the user.

Jussel notes that the product announced this week represents the first FPGA solutions that uses the HTX slot. DRC Computer Corporation has a somewhat similar solution, where its FPGA uses an Opteron socket to directly connect to HyperTransport. As it turns out, DRC is an OEM partner with Celoxica and makes use of the same C compiler technology.

Celoxica's current (beta) customers for the HTX solution are in the financial services, oil & gas, and life sciences industries. With this particular product, users have achieved a 200X performance improvement for the application (offloading a Black Scholes algorithm to the FPGA).

“We've done enough with the finance industry to know that the metric we need to hit is about a 10X price-performance benefit,” says Jussel. “If we hit that 10X factor then it's worth it for them to invest in new technology. And we've been able to show quite a bit greater than 10X for all these applications.”

Even though this HTX product includes the FPGA card, hardware is not Celoxica's main focus. The company's real goal is to be the leader in compilers and the run-time support for FPGAs. Jussel says that compared to other FPGA compiler companies, Celoxica is quite a bit larger and more established, having developed and matured its software technology over the past 10 years. It provides the compiler for the SGI RASC RC100 system as well as Cray FPGA systems, not to mention its large customer base in the embedded computing space — still the majority of their business. But because FPGAs have this unique aspect of reconfigurability and high performance, the company believes that these devices will become ubiquitous throughout computing. And Celoxica wants to be there with their software.

Says Jussel: “We really want to be the Microsoft of FPGA computing. We want to provide the compilers and RTOS for that solution.”

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!

MLCommons Launches New AI Safety Benchmark Initiative

April 16, 2024

MLCommons, organizer of the popular MLPerf benchmarking exercises (training and inference), is starting a new effort to benchmark AI Safety, one of the most pressing needs and hurdles to widespread AI adoption. The sudde Read more…

Quantinuum Reports 99.9% 2-Qubit Gate Fidelity, Caps Eventful 2 Months

April 16, 2024

March and April have been good months for Quantinuum, which today released a blog announcing the ion trap quantum computer specialist has achieved a 99.9% (three nines) two-qubit gate fidelity on its H1 system. The lates Read more…

Mystery Solved: Intel’s Former HPC Chief Now Running Software Engineering Group 

April 15, 2024

Last year, Jeff McVeigh, Intel's readily available leader of the high-performance computing group, suddenly went silent, with no interviews granted or appearances at press conferences.  It led to questions -- what's Read more…

Exciting Updates From Stanford HAI’s Seventh Annual AI Index Report

April 15, 2024

As the AI revolution marches on, it is vital to continually reassess how this technology is reshaping our world. To that end, researchers at Stanford’s Institute for Human-Centered AI (HAI) put out a yearly report to t Read more…

Crossing the Quantum Threshold: The Path to 10,000 Qubits

April 15, 2024

Editor’s Note: Why do qubit count and quality matter? What’s the difference between physical qubits and logical qubits? Quantum computer vendors toss these terms and numbers around as indicators of the strengths of t Read more…

Intel’s Vision Advantage: Chips Are Available Off-the-Shelf

April 11, 2024

The chip market is facing a crisis: chip development is now concentrated in the hands of the few. A confluence of events this week reminded us how few chips are available off the shelf, a concern raised at many recent Read more…

MLCommons Launches New AI Safety Benchmark Initiative

April 16, 2024

MLCommons, organizer of the popular MLPerf benchmarking exercises (training and inference), is starting a new effort to benchmark AI Safety, one of the most pre Read more…

Exciting Updates From Stanford HAI’s Seventh Annual AI Index Report

April 15, 2024

As the AI revolution marches on, it is vital to continually reassess how this technology is reshaping our world. To that end, researchers at Stanford’s Instit Read more…

Intel’s Vision Advantage: Chips Are Available Off-the-Shelf

April 11, 2024

The chip market is facing a crisis: chip development is now concentrated in the hands of the few. A confluence of events this week reminded us how few chips Read more…

The VC View: Quantonation’s Deep Dive into Funding Quantum Start-ups

April 11, 2024

Yesterday Quantonation — which promotes itself as a one-of-a-kind venture capital (VC) company specializing in quantum science and deep physics  — announce Read more…

Nvidia’s GTC Is the New Intel IDF

April 9, 2024

After many years, Nvidia's GPU Technology Conference (GTC) was back in person and has become the conference for those who care about semiconductors and AI. I Read more…

Google Announces Homegrown ARM-based CPUs 

April 9, 2024

Google sprang a surprise at the ongoing Google Next Cloud conference by introducing its own ARM-based CPU called Axion, which will be offered to customers in it Read more…

Computational Chemistry Needs To Be Sustainable, Too

April 8, 2024

A diverse group of computational chemists is encouraging the research community to embrace a sustainable software ecosystem. That's the message behind a recent Read more…

Hyperion Research: Eleven HPC Predictions for 2024

April 4, 2024

HPCwire is happy to announce a new series with Hyperion Research  - a fact-based market research firm focusing on the HPC market. In addition to providing mark 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…

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…

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…

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…

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…

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…

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…

Leading Solution Providers

Contributors

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…

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…

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…

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…

Eyes on the Quantum Prize – D-Wave Says its Time is Now

January 30, 2024

Early quantum computing pioneer D-Wave again asserted – that at least for D-Wave – the commercial quantum era has begun. Speaking at its first in-person Ana Read more…

GenAI Having Major Impact on Data Culture, Survey Says

February 21, 2024

While 2023 was the year of GenAI, the adoption rates for GenAI did not match expectations. Most organizations are continuing to invest in GenAI but are yet to Read more…

Intel’s Xeon General Manager Talks about Server Chips 

January 2, 2024

Intel is talking data-center growth and is done digging graves for its dead enterprise products, including GPUs, storage, and networking products, which fell to Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire