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

Stricter guidelines imposed by version 3 of the IEC standard for harmonic current emissions push designers to embrace power-factor-correction methodologies.
The flyback topology is a versatile, widely used, switched-mode power-supply design with some interesting characteristics that brings performance and BOM advantages to many applications...
Significant complications arise when trying to get your power supply working from a wall outlet.
Power supplies are everywhere and there are a lot of options for designers. Knowing the differences can help in selecting one for a particular application.