Presentation
SaGraph: A Similarity-aware Hardware Accelerator for Temporal Graph Processing
DescriptionAlthough several software/hardware solutionshave been proposed for efficient temporal graph processing, they still suffer from serious irregular data access due to the uncoordinated graph traversal. To overcome these limitations, we present SaGraph, a domain-specific hardware accelerator to support the efficient processing of temporal graph. Specifically, SaGraph can dynamically coordinate the graph traversals and adaptively cache the vertex states to fully exploit the data access similarity of temporal graph processing. We implemented and evaluated SaGraph on a Xilinx Alveo U280 FPGA card. Compared with the cutting-edge software and hardware solutions, SaGraph achieves 8.5x-157.3x, 4.2x-16.1x speedups and 34.7x-423.6x, 5.3x-14.7x energy savings, respectively.
Event Type
Research Manuscript
TimeWednesday, July 12th2:25pm - 2:40pm PDT
Location3006, 3rd Floor
Design
AI/ML Architecture Design