btzq commented on issue #12075: URL: https://github.com/apache/cloudstack/issues/12075#issuecomment-3564014825
@delacosta456 My company actively uses HashiCorp Nomad on top of VMs provisioned by Apache CloudStack, so I can relate to this feature request. Today, one of the practical limitations we face is with autoscaling container workloads. Nomad can easily scale Docker containers within an existing VM, but it has no native way to request additional VMs from CloudStack when cluster capacity is exhausted. This means the VM hosting the Nomad client must be sized generously from the start, otherwise autoscaling is limited to the resources within that single instance. What would be extremely helpful is a tighter integration layer between Nomad and CloudStack — something where: 1. When Nomad identifies the need to scale out, 2. It could trigger CloudStack to provision a new VM (pre-configured with the Nomad client or via userdata), 3. And the new VM would automatically join the Nomad cluster, 4. Allowing Nomad to place new container allocations on that VM as part of the scaling event. This is a pattern commonly seen with Kubernetes Cluster Autoscaler and various cloud providers. Having a similar mechanism for Nomad + CloudStack would greatly improve elasticity and resource efficiency for users running Nomad clusters on CloudStack. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
