BTAS: A Library for Tropical Algebra

TL;DR

This paper introduces BTAS, a GPU-accelerated library for tropical algebra, demonstrating significant performance improvements over CPU implementations in matrix operations and algorithms like Floyd-Warshall.

Contribution

The paper presents a novel GPU-based library for tropical algebra, enabling efficient vector and matrix computations with improved performance.

Findings

BTAS outperforms CPU implementations in speed.

GPU implementation is cost-effective.

Parallel Floyd-Warshall benefits from GPU acceleration.

Abstract

GPUs are dedicated processors used for complex calculations and simulations and they can be effectively used for tropical algebra computations. Tropical algebra is based on max-plus algebra and min-plus algebra. In this paper we proposed and designed a library based on Tropical Algebra which is used to provide standard vector and matrix operations namely Basic Tropical Algebra Subroutines (BTAS). The testing of BTAS library is conducted by implementing the sequential version of Floyd Warshall Algorithm on CPU and furthermore parallel version on GPU. The developed library for tropical algebra delivered extensively better results on a less expensive GPU as compared to the same on CPU.