Dynamic Neural Accelerator

Overview

EdgeCortix Dynamic Neural Accelerator (DNA) is a flexible neural accelerator IP core with run-time reconfigurable interconnects between compute units, achieving exceptional parallelism and efficiency through dynamic grouping. A single core can scale from 1024 MACs to 32768 MACs, running at clock speeds up to 1 GHz. Configuration is done with the EdgeCortix MERA compiler and software framework.

Key Features

  • Proprietary reconfigurable Data Flow architecture
  • Ground breaking efficiency for deep neural network deployment
  • Seamless integration with standard deep
  • Proprietary, automatic, neural network model and hardware accelerator IP co-design
  • Hyper scalable design with custom interconnect
  • Integer 8bit precision optimized
  • DNA-IP available for custom SoC, PCI-E processor
  • learning SW frameworks - Pytorch, Tensorflow or on edge-server deployment with Xilinx ALVEO and Zynq Ultrascale+ FPGAs

Benefits

  • Run-time reconfigurable
    • Efficient AI inference core in FPGAs or SoCs
  • Low latency
    • < 4 msec when under demanding workloads
  • High utilization
    • +80% for concurrent neural network models

Applications

  • Automotive
  • Robotics
  • Smart City
  • Drones
  • Industry 4.0

Technical Specifications

×
Semiconductor IP