Allow users to leverage Google Cloud Spot VMs when deploying their workloads on an OSD cluster.
Here are three key goals for using Spot VMs in OSD on Google Cloud:
- Reduce infrastructure costs: Leverage significantly discounted Spot VM pricing to lower the total cost of ownership for running container workloads, especially for non-critical, short-lived, or fault-tolerant applications.
- Optimize compute utilization for scalable workloads: Enable customers to run large-scale batch jobs, CI/CD pipelines, ML training, and stateless services more economically by scaling out using Spot VMs when capacity is available.
- Provide flexible workload placement options: Allow customers to define mixed-node strategies using On-Demand and Spot VMs to balance cost, availability, and workload criticality through taints, tolerations, and node selectors.
Allow users to leverage Google Cloud Spot VMs when deploying their workloads on an OSD cluster.
Here are three key goals for using Spot VMs in OSD on Google Cloud: