Configuring and Verifying Compute AZ Management on OpenStack

AZs are logical abstractions visible to end users. They are used to partition clouds without knowing the physical infrastructure. Generally, they are used to logically group and isolate resources on compute nodes in a small region (country/region, data center, or rack).

Host Aggregates (HAs) further logically group and isolate resources on top of AZs. For example, you can plan the physical resources with common attributes under the same HA according to the physical hardware configurations of different compute nodes. Or, you can plan several compute nodes under one HA for the same purpose according to your requirements. Such divisions help improve the OpenStack resource utilization.

Perform the following operations on the controller node.

Typical Scenario:

1. Log in to the OpenStack CLI as the user admin.

   1	source /etc/keystone/admin-openrc

2. The OpenStack client provides the function of creating an HA and associating it with an AZ. For example, you create an aggregation, associate the aggregation with an AZ, and use the OpenStack client to add hosts to the aggregation.

   In the hybrid deployment environment, compute nodes of the same type are deployed in the same AZ.

   1 2	nova aggregate-create agg-x86 az-x86 nova aggregate-create agg-arm az-arm

   

3. If you add a compute node to a host aggregate, the compute node will be automatically added to the associated AZ.

   1 2	nova aggregate-add-host agg-x86 x86-compute nova aggregate-add-host agg-arm arm-compute

   

4. List all host aggregates and display information about a specific aggregate to determine the AZ to which the host aggregate belongs.

   1 2	openstack aggregate list --long openstack aggregate show agg-arm