KubeCon + CloudNativeCon North America 2024

As Kubernetes becomes the leading platform for data processing, mastering the deployment and management of Apache Spark on it is crucial. In this presentation, you'll hear from the new maintainers of the Kubeflow Spark Operator project, who will provide an overview of scaling the Spark Operator on Kubernetes, emphasizing best practices to optimize performance and efficiency. Attendees will explore the migration of the Spark Operator repository from Google to Kubeflow, gaining insights into the roadmap and key takeaways. The session will cover strategies for achieving multi-tenancy, managing multiple Spark Operator instances for large-scale deployments, ensuring robust security, and performing seamless upgrades. Participants will learn advanced techniques to maximize their Spark on Kubernetes deployments, making their data processing pipelines more efficient, reliable, and secure. This talk is for Data, ML, DevOps, and MLOps pros to enhance their Spark on Kubernetes skills.