Summary

In this chapter, you learned:

  • OpenStack API resources exist in projects, and projects exist in domains.

  • Domains can refer to external identity systems, and users from one domain can access projects from other domains.

  • OpenStack users and groups are assigned roles which grant them permissions to manage API resources in a project.

  • Projects can also be assigned API resource quotas and compute resource quotas, to prevent workloads from one project from using all capacity of their cluster.

  • OpenStack clusters can be grouped, each as a region, in a larger private cloud.

  • OpenStack clusters can be partitioned into availability zones (AZs) to manage the reliability and resilience of workloads from infrastructure issues.

  • OpenStack compute nodes can be members of multiple host aggregates to schedule workloads to physical servers having better performance or cost for each workload.

  • OpenStack compute nodes can also be partitioned in cells, independent of AZs and host aggregates, to manage the scalability and resilience of OpenStack services themselves.