Gnn 1 609ed7b4e63ea

An FPGA-Based Solution for a Graph Neural Network Accelerator

May 18, 2021
Graph Neural Networks (GNN) drive high demand for compute and memory performance. Software-only based GNN algorithm processing is insufficient to run these workloads. Learn how FPGA-based hardware accelerators overcome GNN processing challenges.

GNN have very high requirements for computing power and memory, and the software implementation of 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 will review the latest GNN algorithms, current status of acceleration technology research and a discussion of FPGA-based GNN acceleration technology.

Sponsored

The advent of USB Type-C marked a turning point in connectivity. This compact, reversible connector has transformed the way we exchange data and power our devices, offering accelerated...
The Same Sky interconnect group carries a comprehensive line of connectors to reduce the burden on the design engineer. Their wide selection of mechanical configurations and simple...
Same Sky Devices' HDMI connectors are available in Type A receptacle versions conforming to the HDMI 2.0 standard. These HDMI Connectors are available in mid-mount SMT, surface...
From data converters to sensing, beginner to advanced – look to the Analog Design Journal for answers to your analog design questions. Some of the industry’s most knowledgeable...