A recent Intel keynote speech noted that the cost of one transistor is less than the cost of the ink in the period dot at the end of a sentence.
Astounding. Semiconductor transistors are one of humankind’s most amazing inventions, yet they’re more plentiful than grains of rice.
 An FPGA-Based Solution for a Graph Neural Network AcceleratorSponsored by Achronix Graph Neural Networks (GNN) drive high demand for compute and memory performance and a software only based implementation of a GNN does not meet performance targets. As a result, there is an urgent need for hardware-based GNN acceleration. While traditional convolutional neural network (CNN) hardware acceleration has many solutions, the hardware acceleration of GNN has not been fully discussed and researched. This white paper reviews the latest GNN algorithms, the current status of acceleration technology research, and discusses FPGA-based GNN acceleration technology. |
