Starting around 2012, artificial intelligence (AI) and machine learning applications began experiencing dramatic expansion into numerous applications and industries. Since then, remarkable progress has been made on a broad range of applications that include computer vision, natural language processing, autonomous robots, self-driving cars, medical imaging, fraud detection, and more. Until now, machine learning training and inference have required high-performance computing infrastructure, which is not always available to all users in need of these applications and so raises barriers to participation.
This white paper describes a currently available Dell EMC compute platform with Red Hat OpenShift container software. Using NVIDIA GPUs to accelerate AI performance, we ran the performance benchmark MLPerf using version 0.6 for training and 0.5 for inference. The results show that this combination of Dell EMC PowerEdge compute, Red Hat OpenShift container software, and NVIDIA GPUs provides a solution that enables the development, deployment, and scaling of AI and machine learning applications.