copyright | lastupdated | keywords | subcollection | ||
---|---|---|---|---|---|
|
2022-06-20 |
DevSecOps, IBM Cloud |
devsecops |
{{site.data.keyword.attribute-definition-list}}
{: #deployment-model}
Deployment models help you to understand your application deployment process across clusters. {: shortdesc}
{: #deployment-model-push}
The push-based deployment model deploys your application on a single cluster. It does not require any operator installation on your cluster.
For a push-based deployment, complete these steps:
- Log in to your Kubernetes or {{site.data.keyword.openshiftlong_notm}} cluster.
- Apply the deployment files by using the default kubectl{: external} tool.
{: #deployment-model-pull}
The pull-based deployment model is suitable for multi-cluster application deployments. This model uses {{site.data.keyword.satellitelong_notm}} Config as the deployment tool for deploying across multiple clusters. For more information, see Understanding {{site.data.keyword.satelliteshort}} Config.
For a pull-based deployment, complete these steps:
- Create or update the required {{site.data.keyword.satelliteshort}} config and their versions for the deployment files.
- Create or update the {{site.data.keyword.satellitelong_notm}} subscriptions with {{site.data.keyword.satelliteshort}} config and the versions that need to be deployed.
- After the subscriptions are updated, the Razee agent that is installed on the clusters pulls the most recent deployments from {{site.data.keyword.satelliteshort}} config and deploys them across the clusters.