High performance, high throughput and technical batch computing are all about scale and speed. Many applications, such as drug discovery and genomics, financial services and image processing require access to a large and diverse set of computing resources on demand. With more and faster computing power, you can convert an idea into a discovery, a hypothesis into a cure or an inspiration into a product. Google Cloud provides customers with flexible, on-demand access to large amounts of cutting edge high-performance resources with Compute Engine.
Batch on GKE is a cloud native solution for managing HPC, HTC and batch workloads in a way that is optimized for virtual cloud resources yet is portable and works on-premises as well. With the introduction of Batch on GKE, we seek to work with the community to define a new way to do batch computing that is cloud optimized, open, standard and portable.
This Beta release focuses on bringing traditional batch scheduler functionality into a cloud-first world: creating and managing queues, submitting jobs, etc. To the end-user Batch on GKE presents a familiar interface that supports well-understood batch concepts, including
- Queues with priorities
- Jobs that can be added to Queues
- Job dependencies
- The ability to easily specify resources required by a job (CPU, memory)
- Support for submitting shell scripts as batch jobs
This repository contains scripts, tools and sample jobs for use with Batch on GKE.
For more information about Batch on GKE, see
- https://cloud.google.com/kubernetes-engine/docs/how-to/batch/managing-clusters
- https://cloud.google.com/kubernetes-engine/docs/concepts/batch
You can go to google-batch-discuss to reach out to Googlers and the community for assistance.