MLPerf Training Benchmark

by   Peter Mattson, et al.

Machine learning is experiencing an explosion of software and hardware solutions, and needs industry-standard performance benchmarks to drive design and enable competitive evaluation. However, machine learning training presents a number of unique challenges to benchmarking that do not exist in other domains: (1) some optimizations that improve training throughput actually increase time to solution, (2) training is stochastic and time to solution has high variance, and (3) the software and hardware systems are so diverse that they cannot be fairly benchmarked with the same binary, code, or even hyperparameters. We present MLPerf, a machine learning benchmark that overcomes these challenges. We quantitatively evaluate the efficacy of MLPerf in driving community progress on performance and scalability across two rounds of results from multiple vendors.


Benchmarking Contemporary Deep Learning Hardware and Frameworks:A Survey of Qualitative Metrics

This paper surveys benchmarking principles, machine learning devices inc...

Benchmarking TinyML Systems: Challenges and Direction

Recent advancements in ultra-low-power machine learning (TinyML) hardwar...

Analysis of DAWNBench, a Time-to-Accuracy Machine Learning Performance Benchmark

The deep learning community has proposed optimizations spanning hardware...

Edge Impulse: An MLOps Platform for Tiny Machine Learning

Edge Impulse is a cloud-based machine learning operations (MLOps) platfo...

AIPerf: Automated machine learning as an AI-HPC benchmark

The plethora of complex artificial intelligence (AI) algorithms and avai...

Chakra: Advancing Performance Benchmarking and Co-design using Standardized Execution Traces

Benchmarking and co-design are essential for driving optimizations and i...

Codabench: Flexible, Easy-to-Use and Reproducible Benchmarking for Everyone

Obtaining standardized crowdsourced benchmark of computational methods i...

Code Repositories


Reference implementations of MLPerf™ training benchmarks

view repo

Please sign up or login with your details

Forgot password? Click here to reset