congatec has introduced its new COM-HPC Client Size C module with AMD Ryzen AI Embedded X100 Series processors. The application-ready aReady.COM conga-HPC/cRX1 has been specifically designed for compute-intensive physical AI applications. Featuring up to 16 AMD “Zen 5” CPU cores, up to 40 compute units of the integrated AMD Radeon RDNA 3.5 GPU, and a dedicated NPU, conga-HPC/cRX1 sets a new performance benchmark for COM-HPC Client modules. Its high, integrated AI and FP32 compute performance eliminates the need for discrete AI accelerator cards in many target applications. Combining outstanding compute performance, extensive I/O bandwidth, and support for industrial temperatures ranging from -40°C to +85°C, the module is ideally suited for physical AI applications that consolidate perception, intelligence, and actuation to execute complex tasks in dynamic operating environments.
The COM-HPC module, part of the AMD Kria AI SOM portfolio, leverages the new AMD Ryzen AI Embedded X100 series with up to 16 “Zen 5” CPU cores for time-critical, sequential workloads such as real-time control, sensor fusion, planning, and decision-making. The integrated GPU delivers up to 59 TOPS of dense INT8 AI inference performance and up to 29,7 TFLOPS of FP32 compute performance, accelerating highly parallel workloads such as perception and object detection in collaborative robots and autonomous vehicles. It also enables the local execution of Large Language Models (LLMs). The dedicated NPU additionally provides up to 50 TOPS of AI performance for always-on AI workloads such as object detection and recognition, speech recognition, and image processing. Based on scalable Computer-on-Module (COM) technology, congatec’s solutions enable modular, rugged embedded platforms with long-term availability that combine low system complexity with highest computing performance.
conga-HPC/cRX1 is available in the COM-HPC Client Size C form factor, measuring
Memory-intensive applications benefit from up to 128 GB of LPDDR5X-8533 unified memory, which enables efficient data sharing between the computer engines. This reduces latency, while simplifying system design. The memory is also soldered to increase the module’s robustness against shock and vibration. The capacity and high bandwidth of the unified memory accelerate general-purpose GPU (GPGPU) workloads, which benefit from high data throughput between memory and compute engines. Additionally, up to 512 GB of optional onboard NVMe storage provides fast data access and short loading times. To connect expansion devices, the module offers up to 24 PCIe Gen4 lanes, enabling straightforward integration of numerous low-lane devices such as industrial Ethernet interfaces, fieldbus adapters, or wireless communication modules. Additional interfaces include 2x 2,5 GbE, up to 4x USB 3.2 Gen2, up to 8x USB 2.0, up to 2x SATA 6 Gb/s, 2x I2C, 2x UART, 12x GPIO, 1x SMBus, and 1x SPI. The module has been developed in accordance with IEC 62443-4-1.
conga-HPC/cRX1’s supported operating systems include Microsoft Windows 11, Windows 11 IoT Enterprise, and Linux. As an application-ready aReady.COM module, conga-HPC/cRX1 is also available preconfigured with licensed ctrlX OS, Ubuntu Pro, or KontronOS operating systems. The aReady.VT option, which features integrated conga-zones hypervisor and virtualisation technologies, enables developers to consolidate multiple workloads – such as real-time control, HMI, AI, and IoT gateway functions – onto a single module. For IIoT connectivity, congatec offers aReady.IOT software building blocks, which include conga-connect for enabling data exchange, remote maintenance, and management of the module, carrier boards, and peripherals as needed.
| Tel: | +27 11 319 8600 |
| Email: | avnet-silica-sales-southafrica@avnet.eu |
| www: | www.avnet.com/wps/portal/silica |
| Articles: | More information and articles about Avnet Silica |
© Technews Publishing (Pty) Ltd | All Rights Reserved
printer friendly version