Uploaded October 2022 | Updated September 2026, 11 hours ago
Ray is rapidly gaining momentum as a parallel computing platform that provides a scale-out cluster model inspired by tools such as Spark and Flink, yet also supports a lightweight scale-to-zero Serverless style workflow designed natively for modern container platforms in the Kubernetes ecosystem.
Ray implements a constellation of tools that support data science devops activities ranging from ETL and feature extraction, model training, ML pipelines, through serverless inferencing.
In this talk, Erik Erlandson describes ongoing projects at Emerging Technologies to deploy Ray on OpenShift, integrate it with Open Data Hub and evaluate Ray’s components for Data Science Workflows. He demonstrates Ray in action to run an end to end data science project on OpenShift. The audience learns how to leverage the capabilities of Ray and OpenShift to power their cloud native Data Science workflows.
Ray is rapidly gaining momentum as a parallel computing platform that provides a scale-out cluster model inspired by tools such as Spark and Flink, yet also supports a lightweight scale-to-zero Serverless style workflow designed natively for modern container platforms in the Kubernetes ecosystem.
Ray implements a constellation of tools that support data science devops activities ranging from ETL and feature extraction, model training, ML pipelines, through serverless inferencing.
In this talk, Erik Erlandson describes ongoing projects at Emerging Technologies to deploy Ray on OpenShift, integrate it with Open Data Hub and evaluate Ray’s components for Data Science Workflows. He demonstrates Ray in action to run an end to end data science project on OpenShift. The audience learns how to leverage the capabilities of Ray and OpenShift to power their cloud native Data Science workflows.
