Analyzing GPU Tensor Core Potential for Fast Reductions
Roberto Carrasco, Raimundo Vega, Cristóbal A. Navarro · 2018
The Nvidia GPU architecture has introduced new computing elements such as the tensor cores, which are special processing units dedicated to perform fast matrix-multiplyaccumulate (MMA) operations and accelerate Deep Learning applications. In this work we present the idea of using tensor cores for a different purpose such as the parallel arithmetic reduction problem, and propose a new GPU tensor-core based algorithm as well as analyze its potential performance benefits in comparison to a traditional GPU-based one. The proposed method, encodes the reduction of n numbers as a set of m × m MMA tensor-core operations (for Nvidia's Volta architecture m = 16) and takes advantage from the fact that each MMA operation takes just one GPU cycle. When analyzing the cost under a simplified GPU computing model, the result is that the new algorithm manages to reduce a problem of n numbers in T(n) = 5 log(m2) (n) steps with a speedup of S = 4/5 log2(m2).