Summary: This post details a solution for distributed deep learning training for a High Energy Physics use case, deployed using cloud resources and Kubernetes. You will find the results for training using CPU and GPU nodes. This post also describes an experimental tool that we developed, TF-Spawner, and how we used it to run distributed TensorFlow on a Kubernetes cluster.
The views expressed in this blog are those of the authors and cannot be regarded as representing CERN’s official position.
CERN update, Quantum Diaries, Careers at CERN
Christian Antognini, Karl Arao, Martin Bach, Mark Bobak, Wolfgang Breitling, Doug Burns, Kevin Closson, Cloudera blog, Wim Coekaerts, Bertrand Drouvot, Enkitec blog, Pete Finnigan, Richard Foote, Randolf Geist, Marco Gralike, Brendan Gregg, Kyle Hailey, Tim Hall, Uwe Hesse, Frits Hoogland, Hortonworks blog, Integrity Oracle Security, Tom Kyte, Adam Leventhal, Jonathan Lewis, Cary Millsap, James Morle, Karen Morton, Arup Nanda, Mogens Nørgaard, Oracle The Data Warehouse insider, Oracle Enterprise Manager, Oracle Linux blog, Oracle Multitenant, Oracle Optimizer blog, Oracle R technologies, Oracle Upgrade blog, Oracle Virtualization blog, Kerry Osborne, Tanel Poder, Planet PostgreSQL, Kellyn Pot'Vin, Pythian blog, Greg Rahn, Mark Rittman, Riyaj Shamsudeen, Chen Shapira, Carlos Sierra, Szymon Skorupinski