Accelerating Force-Directed Graph Drawing with RT Cores

TL;DR

This paper introduces a novel method for accelerating force-directed graph drawing by leveraging NVIDIA RT cores, achieving significant speedups over traditional CUDA implementations through hardware-accelerated hierarchical tree traversal.

Contribution

It maps force-directed graph layout computation to a ray tracing problem, enabling hardware acceleration with RT cores for faster graph drawing.

Findings

Achieves 4x to 13x speedup over CUDA implementations.

Demonstrates effective hardware mapping of graph layout algorithms.

Shows potential for real-time large graph visualization.

Abstract

Graph drawing with spring embedders employs a V x V computation phase over the graph's vertex set to compute repulsive forces. Here, the efficacy of forces diminishes with distance: a vertex can effectively only influence other vertices in a certain radius around its position. Therefore, the algorithm lends itself to an implementation using search data structures to reduce the runtime complexity. NVIDIA RT cores implement hierarchical tree traversal in hardware. We show how to map the problem of finding graph layouts with force-directed methods to a ray tracing problem that can subsequently be implemented with dedicated ray tracing hardware. With that, we observe speedups of 4x to 13x over a CUDA software implementation.