Versal AI Edge ACAP drives up ML compute and adds memory for AI

Versal AI Edge chipsets are designed to meet the computing densities required by automated systems with activated intelligence for real-time control. They have 10 times the computing density of Zynq UltraScale + MPSoCs for low latency, intelligent, autonomous systems and support safety standards IEC 61508, DO-254/178 and ISA 26262 for industrial applications, avionics and automotive.

“Applied computing applications require an architecture that can be developed to meet new requirements and scenarios with a combination of flexible computational processing within stringent thermal and latent constraints,” said Sumit Shah, senior director, product management and marketing at Xilinx.

The Versal AI Edge series is based on the 7nm Versal architecture, miniaturized for AI calculations with low latency and operation up to 6W. There are scalar motors for embedded computing, adaptive motors for sensor synthesis and hardware adaptation, as well as intelligent AI output motors that scale to 479 (INT4) TOPS. Xilinx says the figure is “incomparable to ASSPs and GPUs targeting end-to-end applications.” Computing performance is also suitable for advanced loads for processing signals for vision, radar, lidar and software-defined radio.

For connectivity there is LPDDR-4266, 32Gbit / s transceivers, 40G multi-speed Ethernet, PCIe Gen4 with CCIX and built-in MIPI support for visual sensors up to 8Mpixels.

Development tools include Vivado for hardware developers, Vitis’ unified software platform for software developers, Vitis AI for data researchers, and domain-specific operating systems, frameworks, and acceleration libraries for targeted applications.

Design documentation and support are now available for the newest members of the ACAP portfolio for early access customers. Deliveries are expected to begin in the first half of 2022. Then the company will publish a roadmap for automotive and defense devices.