Concept: Taiwanese technology company ADLINK has unveiled an embedded MXM graphics module based on Nvidia Ampere architecture for accelerated computing and AI workloads at the edge. The company claims that the new product offers real-time ray tracing, AI-accelerated graphics, and energy-efficient AI inference acceleration in the compact mobile PCI express (MXM) form factor. The modules can improve responsiveness, precision, and reliability for applications in healthcare, manufacturing, transportation, gaming, and other sectors.
Nature of Disruption: ADLINK’s embedded mobile PCI express module (MXM) graphics module is based on NVIDIA Ampere architecture which enables general computation, image processing and reconstruction, and AI inference at the edge. The company claims that the architecture combines RT Cores, Tensor Cores, CUDA Cores, PCIe Gen 4, and NVIDIA video codecs to enhance the performance of the module and offer more features. The novel module offers up to 5,120 CUDA Cores, 160 Tensor Cores, and 40 RT Cores with support for PCIe Gen 4 and up to 16GB GDDR6 memory at up to 115 watts of TGP. It is compact than PCI Express graphics cards and can operate under extreme temperature, shock, and vibration, and has corrosion resistance, as claimed.
Outlook: Technology players are shifting computing and AI workloads from the cloud to the edge to reduce response time, improve security, and lower communication costs. This can increase the demand for modules with increased performance and power efficiency. The new embedded MXM GPU module has several applications such as accelerated image reconstruction for mobile X-ray, ultrasound, and endoscopic systems, real-time object detection on railways or airport runways to enhance transport safety, and navigation and route planning for autonomous drones and mobile robots to assist with last-mile delivery.