CPU Tech Brings Real-Time Modeling to Electronics Design

By Nicole Hemsoth

April 11, 2008

A little less than a year ago, CPU Tech announced a collaborative effort with BAE Systems (www.baesystems.com) to begin development of a virtual model of the Bradley Combat Systems vehicle. The announcement stated that networked engineering teams would be able to simulate and validate complete electronic systems including actual Bradley software — running at real-time speeds, providing unprecedented visibility into the complete electronics of the entire vehicle — a modeling and simulation feat that had never been done before. As quoted in that announcement, BAE was the first corporation in the world to adopt high-fidelity real-time virtual system development on a major platform, and referred to this new virtual capability as a “technology breakthrough.”

In this interview, Ed King, CPU Tech founder and CEO, talks about the real-time simulation work his company is doing with BAE Systems, and the nature of the technology behind it.

HPCwire: What has transpired since the announcement and what has CPU Tech learned from this collaborative effort?

Ed King: The collaborative announcement with BAE was a validation of our efforts over the last 15 years and a substantiation of our announcement of the SystemLab PS in May of 2007. BAE is just the first company to realize the faster time to market benefit of virtual prototyping through real-time simulation and validation of complex electronic systems with extraordinary visibility into every wire, every memory, every processor and every line of code.

We’ve achieved the milestones identified for this time period and are defining new milestones and new projects based on our initial successes. Modeling the Bradley has convinced us that our technology will have a tremendously positive impact on advanced vehicle design in numerous industries, regardless of their complexity. There are over 30 discrete electronic systems in the Bradley and each one of these systems is complex by itself. When it comes to modeling and simulation for vehicle electronics, we started with perhaps the most complex and challenging platform out there.

Let me explain what I mean. One of the most difficult challenges for developers of these extremely complex vehicles is making sure that all of these systems can work together as a cohesive platform. SystemLab PS was designed to address this issue by providing simultaneous real-time simulation of all of the vehicle’s electronic systems, as well as comprehensive visibility and control across the entire vehicle’s electronic systems, including platform architecture and analysis, hardware / software development and debug, diagnostic test development, and fault insertion. We are continuing to work with BAE to apply this next generation of modeling and simulation technology to various derivatives of the Bradley platform, and, moving forward, we hope to have the opportunity to work with BAE on the more than 40 different platforms similar to the Bradley.

We are in discussions with a number of other companies to model their large scale platforms as well. We see this as a substantial source of business growth for CPU Tech; both in the modeling of the systems and in the delivery of the SystemLab hardware and software solutions. And, we’ve started discussions with several automotive companies regarding the benefits SystemLab offers in the area of new vehicle development and design.

HPCwire: So, when you talk about modeling and simulation, I’m sure many of our readers think of physical models — such as CAD system drawings — depicted in the videos we see in many of the booths at the annual supercomputing show demonstrating vehicle crash dynamics. How would you describe the difference or, how would you draw a comparison, between using a CAD system to model the physical characteristics of a vehicle and what can now be done with the SystemLab?

King: That’s a great question. The SystemLab PS does for the design of the vehicle electronics what using a leading-edge 3-D CAD system to model and simulate the physical characteristics of a vehicle does for the mechanical design. In both cases, you can operate the vehicle in the virtual domain and observe its operation. You can also simulate what will happen as you change the circumstances of the operation. For a 3-D model you can simulate what happens when a car hits something. For a SystemLab PS model you can simulate what happens when there is an electronic failure in either the software or the hardware.

They both are general purpose in that they become whatever model is loaded. In fact, it is possible to integrate SystemLab PS with a 3-D CAD system to create a total virtual vehicle where changes in one model could cause a change in the other. Pushing the button to put down the window would cause the virtual electronics to roll down the window. Hitting something with the virtual car would cause the electronics to deploy the airbags just as it would in a real car.

There are big differences between traditional electronic design tools and SystemLab — namely performance and fidelity. There are ways to model large scale electronics today; in a very high-level, abstract manner which gives you a rough sense of what may work and what may not work. However, it is very difficult to validate actual software running with this high level modeling and it is impossible do it anywhere near real-time speeds.

The SystemLab PS model gives you the fidelity and accuracy of the design and lets you simulate and validate it in real-time. The model gives engineers the ability to look into any part of the model during the simulation and control the simulation down to stepping through lines of code on the processors. Think about this — the model is the design!

HPCwire: What would you list as the three biggest advantages — or benefits — to vehicle engineering teams adopting this real-time modeling and simulation capability? Why is it so important?

King: There are actually a number of very distinct benefits that organizations will realize. First, and of course the one that every company whose revenue depends on bringing a vehicle to market will focus on, is the speed improvement — or time reduction in terms of time-to-solution. In a competitive market, this factor will drive a company’s success or failure in terms of capturing market share.

Next, would be what several customers have referred to as a breakthrough in the visibility now possible into various parts of the design. As it turns out, this is one of the biggest issues with hardware or high-level models and an area where the designers have struggled — having to work in a piece meal fashion without the benefit of seeing how all systems interact.

And finally, if you want to make sure everything works before you build it, and this is where I like to use the phrase, “the model is the design” — the accuracy and the fidelity of the simulation offered by the SystemLab Platform Simulator simply has not been available previously. It’s a tremendous tool and will have a significant impact on the design and development process of vehicles.

HPCwire:  The opportunities for a company to bring a truly “breakthrough” technology to market are few and far between. What makes this a breakthrough technology, and how has CPU Tech been able to accomplish this?

King: The breakthrough of the SystemLab PS is a result of 15 years of work and dedication to solve our own design challenges. We have been architecting and designing complex systems for many years and have seen the same problems that BAE has seen. The SystemLab software development began early and quickly proved itself as a multi-phase design environment. Our own engineers used it to do high performance gate-level, schematic-level and system-level design entry and simulation. In 1993 we simulated a PC, at the gate level, booting Windows.

As we developed our Acalis family of field programmable multi-core (FPMC) Systems-on-Chips (SoCs), we realized that we could combine them with our software and some “special sauce” — aka design techniques — to make the simulations run at real-time speeds. The extraordinary visibility came from our patented Real-Time Non-intrusive Interface which lets you see inside the cores of the SoCs.

Once we had a virtual simulation environment, it was natural to develop the regression environment to allow you to simulate many different tests, sometimes overnight, and record their success or failure. It was also easy to virtually inject faults, such as failed devices or broken wires, into a design and see how it responds. Some faults are very difficult and/or very costly to produce in a real hardware system.

So, born out of our own needs and our desire to find a more efficient process, the SystemLab environment evolved. We’re very fortunate that everything came together the way it did. No one else can simulate the entire electronics suite of extremely large and complex platforms made up of multiple systems, much less do it in real time with platform-wide visibility and control. The combination of these capabilities enables fundamental changes in how complex systems and platforms are developed and managed. True hardware / software co-development is now a reality, even for embedded, real-time systems.

HPCwire: What about competition; what are the alternatives for your solution?

King: You have to remember — the large scale, high-fidelity, real-time electronics simulation market is new. Some companies can simulate electronic systems, but they do it very slowly and only on small systems such as chips or boards. Other companies say they do it more quickly at high levels of abstraction, but rarely come close to real time speeds.

The alternatives are quite varied. Some hardware developers will perform a combination of simulation and emulation at the chip and board level using an assortment of third party and home-grown tools and methods. These approaches suffer from extremely limited speeds, constraining simulation to only seconds. These methods also result in inaccurate timing, which in turn results in bugs missed during emulation. As soon as the basic functionality has been established, a hardware prototype of the system is built.

For a complex system, the time it takes to get to this point is measured in years. Only after the hardware prototype is available can the software debugging begin. Once the system hardware and the software have been debugged, then it is integrated, that is, connected to the other systems in the platform.

The integration is performed in a System Integration Lab (SIL). These are expensive and scarce setups of the prototype electronics hardware, hooked up together as it would be in a real vehicle or platform. The software developers get to line up to test their software, often only for a few hours once every few weeks. When their software doesn’t work, they often have no idea where to begin to look for the cause of failure. Sometimes it’s their software; sometimes it’s someone else’s software; sometimes it’s a failed electronic component; and sometimes it’s incorrect system logic.

These SILs are in use today and strangling the drive to develop new platforms and add new capabilities to existing systems. Boeing recently cited system integration difficulties as one of the reasons for its 787 delays.

SystemLab PS provides an affordable, scalable, and configurable, virtual SIL with visibility into every wire, every memory, every processor and every line of code. When models are updated, they can be “published” to all SystemLab PS virtual SILs electronically. This update is very expensive and logistically challenging for actual hardware SILs.

With SystemLab PS, the model is the design. If the model works, the design will work.

HPCwire: How large is CPU Tech?

King: CPU Tech is a small company. It was founded in 1989 with steady growth since then. We provide technology, tools and services to customers who design and integrate complex systems. We partner with IBM for the secure and trusted fabrication of our custom System-On-Chip devices. Our customers include IBM, AMD, TI, NCR, BAE Systems, Honeywell, Lockheed Martin, U.S. Navy and many others, and we hope to announce several automotive industry clients later this year.

HPCwire: What is the typical timeframe from the day when someone places an order with you to the day when the SystemLab Platform Simulator goes live in a production environment?

King: This depends on what is modeled — how complex it is — and who does the modeling. When we do the modeling of a platform, it can take anywhere from three to nine months. As our customers get trained, their productivity should match or exceed ours.

HPCwire: Is CPU Tech profitable?

King: Yes. CPU Tech has grown at an average of over 20 percent per year since its inception in 1989. The company was entirely self-financed until the Carlyle Group made its 25 percent investment. With several of our systems in production, several modeling contracts for our SystemLab PS and the second generation of our Acalis family of Field Programmable Multi-Core chips this year, the future looks good.

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