Benchmarking Deep Neural Network Inference Performance on Serverless Environments With MLPerf

Unai Elordi, Luis Unzueta, Jon Goenetxea, Sergio Sanchez-Carballido, Ignacio Arganda-Carreras, Oihana Otaegui

Abstract: We provide a novel decomposition methodology from the current MLPerf benchmark to the serverless function execution model. We have tested our approach in Amazon Lambda to benchmark the processing capabilities of OpenCV and OpenVINO inference engines.

External IDs:doi:10.1109/ms.2020.3030199