Upgrade Management

ModelArts Real-Time Service Upgrade

For a deployed service, you can change the AI application version to upgrade it.

Services can be upgraded in three modes: full upgrade, rolling upgrade (increase instances), and rolling upgrade (decrease instances). For details about the three upgrade modes, see Figure 1.

• Full upgrade

  Resources that are twice as many as those used by the service will be used to create new-version instances in full mode.

• Rolling upgrade (increase instances)

  Extra resources than those used by the service will be used for a rolling upgrade. A larger number of instances to be increased will lead to a faster upgrade.

• Rolling upgrade (decrease instances)

  Certain nodes that were intended to run services will be used for a rolling upgrade. A larger number of instances to be reduced will lead to a faster upgrade but a higher probability of service interruption.

Figure 1 Service upgrade process

For details about how to upgrade an inference service, see Upgrading a Service.

Image Upgrade

ModelArts provides three function modules: DevEnviron, training management, and inference deployment. The three modules provide base images by the same process. These images are upgraded irregularly to fix vulnerabilities.