ScalaBFS: A Scalable BFS Accelerator on HBM-Enhanced FPGAs

TL;DR

ScalaBFS is a scalable FPGA-based BFS accelerator that leverages HBM memory to achieve near-linear performance scaling and high throughput comparable to top GPU solutions.

Contribution

It introduces a scalable FPGA-based BFS accelerator that exploits HBM channels for high performance and demonstrates near-linear scaling on real hardware.

Findings

Achieves up to 19.7 GTEPS on U280 FPGA with 32 PCs and 64 PEs.

Performance scales almost linearly with available HBM memory channels.

Comparable BFS throughput to Nvidia V100 GPU on real-world graphs.

Abstract

High Bandwidth Memory (HBM) provides massive aggregated memory bandwidth by exposing multiple memory channels to the processing units. To achieve high performance, an accelerator built on top of an FPGA configured with HBM (i.e., FPGA-HBM platform) needs to scale its performance according to the available memory channels. In this paper, we propose an accelerator for BFS (Breadth-First Search) algorithm, named as ScalaBFS, that builds multiple processing elements to sufficiently exploit the high bandwidth of HBM to improve efficiency. We implement the prototype system of ScalaBFS and conduct BFS in both real-world and synthetic scale-free graphs on Xilinx Alveo U280 FPGA card real hardware. The experimental results show that ScalaBFS scales its performance almost linearly according to the available memory pseudo channels (PCs) from the HBM2 subsystem of U280. By fully using the 32 PCs and building 64 processing elements (PEs) on U280, ScalaBFS achieves a performance up to 19.7 GTEPS (Giga Traversed Edges Per Second). When conducting BFS in sparse real-world graphs, ScalaBFS achieves equivalent GTEPS to Gunrock running on the state-of-art Nvidia V100 GPU that features 64-PC HBM2 (twice memory bandwidth than U280).