TY - JOUR T1 - MLPerf: An industry standard benchmark suite for machine learning performance JF - IEEE Micro Y1 - 2020 A1 - Mattson Peter A1 - Vijay Reddi A1 - Cheng, Christine A1 - Coleman, Cody A1 - Diamos, Greg A1 - Kanter, David A1 - Micikevicius, Paulius A1 - Patterson, David A1 - Schmuelling, Guenther A1 - Tang, Hanlin A1 - Gu-Yeon Wei A1 - Carole-Jean Wu AB - In this article, we describe the design choices behind MLPerf, a machine learning performance benchmark that has become an industry standard. The first two rounds of the MLPerf Training benchmark helped drive improvements to software-stack performance and scalability, showing a 1.3× speedup in the top 16-chip results despite higher quality targets and a 5.5× increase in system scale. The first round of MLPerf Inference received over 500 benchmark results from 14 different organizations, showing growing adoption. PB - IEEE VL - 40 UR - https://doi.org/10.1109/MM.2020.2974843 IS - 2 ER -