Strategies for managing your HPC cluster in the Cloud

If you have a long-running autoscaling HPC cluster on Google Cloud Platform, infrastructure as code and continuous integration can help you simplify management of your cloud resources. Infrastructure-as-code allows you to version control all of your cloud resources including IAM policies, networking and firewall rules, and your HPC cluster resources including partitions and even which images you are using. In this livestream, we'll show you how to easily set up a Google Source Repository to manage your HPC cluster resources on Google Cloud Platform using a combination of Google Cloud Build, Packer, and Terraform. We'll share with you a few publicly available resources on Github that can help you quickly get started with managing your cluster. You will also learn about an ideal autoscaling HPC cluster setup that will allow you to easily incorporate new image releases from Fluid Numerics or from your own organization's custom VM image repository.

Github resources