This year PICMG celebrates thirty years of developing open computing specifications. And despite three decades of open hardware specs that are used by thousands of companies and countless engineers worldwide, the organization is still largely an unknown—even in our own industry.
But from behind the scenes, PICMG is responsible for billions of dollars of business. It has reduced time to market in virtually every electronics-driven industry. It has empowered companies to innovate by adopting off-the-shelf technologies, giving them space to focus on their core competencies. It has built markets based on coopetition, where companies collaboratively develop open, interoperable specifications then go toe-to-toe once they are ratified.
Even less recognized are the individuals who donate hundreds if not thousands of hours to the creation of PICMG specifications. These engineers are rarely acknowledged for their contributions. They are truly unsung heroes, and PICMG specifications wouldn’t exist without them.
One of these unsung heroes is Stefan Milnor, who recently retired from his role as VP of Engineering at Kontron. In parallel, he stepped down as the long-time editor of COM Express and COM-HPC specifications.
Stefan has been involved in PICMG since the beginnings of COM Express, which is undeniably the most successful computer-on-module in a billion-dollar COMs market. As editor, Stefan incorporated input from numerous technical subcommittees over the years and implemented it into specifications that have been adopted by thousands of organizations building embedded systems. It’s a difficult job that requires a rare mix of technical acumen and attention to detail, and his skill and efficiency will be missed by us all—including some who never knew he was the hand behind their downloaded spec.
Stefan has always been a very private individual. In fact, by the time we could reach out to him for comment he had already left. In some ways, it’s a fitting conclusion for someone who performed yeoman’s work for decades without reward. And although he probably isn’t reading this, his contributions to PICMG and the embedded computing industry deserve to be acknowledged.
Thank you, Stefan, for helping make PICMG what it is today: 30 years strong and counting.
— Jessica Isquith, President, PICMG
Admin Note: Contact Doug Sandy, PICMG CTO, at do**@pi***.org to learn how you can get involved in PICMG’s technical working groups. Contact me, je**@pi***.org if you are interested in joining PICMG or have any questions about our organization.
The Members Only interview series highlights leaders from within PICMG and throughout the open standards development community. We recognize their contributions and seek insight into their thought processes and strategies driving open technology-powered industries forward.
This issue we introduce Jens Hagemayer, a research associate at the Bielefeld University. Jens and his team have been intimately involved in the development of COM-HPC since its inception, championing the use of heterogeneous modules designed around FPGAs. Now they are investigating ways that Compute Express Link (CXL) specifications can take COM-HPC into new use cases.
PICMG: Can you describe your work outside of PICMG as well as the role you played in development of the COM-HPC specification?
JENS: I am currently engaged in research at Bielefeld University, focusing on the development of heterogeneous and reconfigurable computing technologies for a wide range of applications. These include the Internet of Things (IoT), edge computing, cloud computing, and high-performance computing (HPC).
My involvement with the early stages of the COM-HPC specification centered on leading the development of the Platform Management Interface Specification and expanding the Embedded EEPROM Specification for COM-HPC.
PICMG: You recently raised awareness about the CXL standard within the COM-HPC community. What is CXL and why is it relevant for PICMG COM-HPC developers and users?
JENS: CXL, or Compute Express Link, is a high-speed, high-capacity interconnect standard that facilitates efficient communication between CPUs, memory, and peripherals using the PCIe physical layer. Its support for cache coherency, disaggregation, and scalable architectures makes it a compelling choice for modular form factors like COM-HPC, driving its popularity among developers and users seeking advanced computing solutions.
PICMG: Why is CXL 3.1 significant in the context of COM-HPC? What use cases or capabilities will it drive in the COM-HPC ecosystem?
JENS: CXL introduces features that cater to the demanding requirements of cloud and high-performance computing systems. Its emphasis on scalable architectures, disaggregation, and cache coherency is particularly relevant for COM-HPC, offering the potential to revolutionize the way modular computing platforms are designed and utilized.
The integration of CXL into COM-HPC could facilitate the development of more sophisticated computing solutions, enabling the seamless coupling of specialized accelerators and the establishment of cache-coherent multi-socket systems. These advancements promise to unlock new possibilities for COM-HPC applications, ranging from data-intensive analytics to AI and machine learning workloads, driving innovation in modular computing technologies.
PICMG: Given that CXL targets PCIe, has it been compatible with COM-HPC to date?
JENS: The relationship between CXL and COM-HPC is fundamentally influenced by CXL’s reliance on the PCIe physical layer for connectivity. This means that while direct compatibility between previous versions of CXL and COM-HPC has not been explicitly defined, the architectural underpinnings allow for potential integration.
The absence of CXL in the current COM-HPC specification, coupled with the lack of support in existing modules, suggests that the integration of CXL represents a forward-looking opportunity for enhancing COM-HPC. Such integration is anticipated to require minimal modifications to the specification, paving the way for future advancements in modular computing.
PICMG: What does the COM-HPC community need to know about the CXL market or technical requirements to capitalize on the opportunity?
JENS: To fully leverage the potential that CXL brings to the COM-HPC community, it is crucial to understand the intricacies of CXL’s market dynamics and technical specifications. This involves a deep dive into the architecture of CXL, including its device types—such as Type 1 for I/O devices, Type 2 for cache-coherent devices, and Type 3 for memory expander devices. Additionally, understanding the topology options that CXL supports, including switch-based topologies for larger, more complex systems, can empower developers to design COM-HPC solutions that are both innovative and future-proof.
Staying abreast of the evolving CXL specifications and market trends will enable the COM-HPC community to identify new opportunities for integration and application, ensuring that COM-HPC modules remain at the forefront of technological advancement.
PICMG: What are you and Bielefeld University doing with respect to CXL today?
JENS: We are working on integrating CXL within the RISC-V ecosystem, a venture that holds promising implications for the future of computing architectures. Our work focuses on the development of innovative bridge technologies that facilitate communication between the RISC-V Coherent Hub Interface (CHI) and CXL, using FPGA-based modules, which we refer to as microservers.
This endeavor is not just about bridging two technical standards; it’s about creating a foundation for next-generation computing platforms that can seamlessly integrate diverse processing and memory resources. By developing these bridges, we aim to enable more efficient, scalable, and flexible computing architectures that can cater to the demanding requirements of modern applications, ranging from AI and machine learning to big data analytics.
PICMG: Where can interested parties go to find more information on CXL?
JENS: The CXL Consortium website serves as the primary repository of knowledge. This platform not only provides access to the official CXL specifications and technical documents, but also offers insights into the latest developments, industry adoption stories, and educational resources.
Deggendorf, Germany, 20 December 2023 * * * congatec – a leading vendor of embedded and edge computing technology – is introducing its latest range of COM Express Compact modules based on the Intel® Core™ Ultra processors. Providing a unique combination of heterogeneous compute engines, including CPU, GPU and NPU, the new modules are an ideal fit to run demanding AI workloads at the edge.
Next to the powerful P-cores and efficient E-cores for general computing and the high-performance Intel® Arc™ GPU for graphics-intensive tasks, the integrated Neural Processing Unit (NPU) called Intel® AI Boost contributes advanced neural processing capabilities to the overall computational architecture. The integrated NPU enables highly efficient integration of advanced artificial intelligence workloads at lower system complexity and costs than discrete accelerators. This makes the new Intel Core Ultra processor-based Computer-on-Modules especially beneficial for combining high-performance real-time computing with powerful AI capabilities in surgery robots and medical imaging and diagnostic systems, where automatically generated critical findings can support medical personnel. Other application targets are situational awareness in industrial applications such as inspection systems, stationary robotic arms, autonomous mobile robots (AMRs), and autonomous guided vehicles (AGVs), to name just a few.
“The new conga-TC700 COM Express Compact Computer-on-Modules provide application-ready AI capabilities in a plug-and-play form factor. Its ecosystem of customer solution-focused products and services significantly improves the time-to-market for implementing the latest state-of-the-art x86 with powerful AI capabilities as required in industrial process control, microgrid controller, medical ultrasound and x-ray, automated check-out terminals, powerful AMRs, and many more,” says Tim Henrichs, VP Global Marketing & Business Development at congatec. “OEMs can simply upgrade existing applications by exchanging the module and instantly gain access to cutting-edge AI technologies. It has never been easier to integrate artificial intelligence in x86-based systems.”
The feature set at a glance
The new conga-TC700 COM Express Compact Computer-on-Modules with Intel Core Ultra processors (code named Meteor Lake) are among the most power-efficient x86 client SoCs available on the market. Up to 6 P-Cores, up to 8 E-Cores, and 2 Low Power E-Cores support up to 22 threads, making it possible to consolidate distributed devices onto a single platform for the lowest total cost of ownership. The SoC-integrated Intel Arc GPU with up to 8 Xe Cores and up to 128 EUs can handle stunning graphics up to 2x 8K resolution and ultra-fast GPGPU-based vision data (pre)processing. The integrated NPU Intel AI Boost executes machine learning algorithms and AI inferences particularly efficient. Up to 96 GB DDR SO-DIMM with in-band ECC at 5600 MT/s contributes to power-efficient high data throughput and low latency.
The modules are supported by congatec’s OEM solution-focused high-performance ecosystem, including highly efficient active and passive cooling solutions and ready-to-use evaluation carrier boards. Customers can order the modules with pre-evaluated real-time hypervisor technology from Real-Time Systems for virtual machine deployments and workload consolidation in edge computing scenarios. Service offerings include shock and vibration tests for custom system designs, temperature screening, and high-speed signal compliance testing, along with design-in services and all required training sessions to simplify the use of congatec’s embedded computer technologies to round off the ecosystem.
The new conga-TC700 COM Express Compact Type 6 modules support the embedded temperature range from 0 °C to 60 °C and are available in the following standard configurations:
congatec is a rapidly growing technology company focusing on embedded and edge computing products and services. The high-performance computer modules are used in a wide range of applications and devices in industrial automation, medical technology, robotics, telecommunications and many other verticals. Backed by controlling shareholder DBAG Fund VIII, a German midmarket fund focusing on growing industrial businesses, congatec has the financing and M&A experience to take advantage of these expanding market opportunities. congatec is the global market leader in the computer-on-modules segment with an excellent customer base from start-ups to international blue chip companies. More information is available on our website at www.congatec.com or via LinkedIn, X (Twitter) and YouTube.
