SC14 New Orleans, LA

The International Conference for High Performance Computing, Networking, Storage and Analysis

A Roofline Performance Analysis of an Algebraic Multigrid Solver.

Authors: Alex Druinsky (Lawrence Berkeley National Laboratory), Brian Austin (Lawrence Berkeley National Laboratory), Xiaoye S. Li (Lawrence Berkeley National Laboratory), Osni Marques (Lawrence Berkeley National Laboratory), Eric Roman (Lawrence Berkeley National Laboratory), Samuel Williams (Lawrence Berkeley National Laboratory)

Abstract: We present a performance analysis of a novel element-based algebraic multigrid (AMGe) method combined with a robust coarse-grid solution technique based on HSS low-rank sparse factorization. Our test datasets come from the SPE Comparative Solution Project for oil reservoir simulations. The current performance study focuses on one multicore node and on bound analysis using the roofline technique. We found that keeping a small value of spectral tolerance is most critical to achieve the best AMG solver performance. Our roofline bound estimate is within 23% accuracy compared to the actual runtime on a Cray XC30 12-cores processor.

Poster: pdf
Two-page extended abstract: pdf

Poster Index