Deployment and component profiles
Decide on whether to deploy three replica or one replica API Connect subsystems, and the memory and CPU resources to allocate to the subsystems.
API Connect is composed of
three components: Management,
Portal, and Analytics, and a supporting program: the DataPower Gateway, see API Connect components.
Note: Throughout this document, the DataPower Gateway supporting program is also
referred to as a
The Management, Portal, Analytics, and
Gateway each have a set of capacity and HA profiles, called component profiles. OpenShift and Cloud Pak for Integration also have API Connect deployment
profiles that
define the component
profiles that are used by each component of a complete API Connect deployment.componentor a
subsystem. The Management, Portal, and Analytics components are also referred to as
subsystems.
The naming convention of the profiles is
nAxcB.mC
where:- nA indicates the number (A) of replicas of each pod, where each replica must run on a different worker node.
- cB indicates the minimum number of virtual cores (B) required on the worker node to deploy the profile.
- mC indicates the minimum amount of memory (C) in GB required by the profile on the worker node.
Component profiles
Each API Connect component has a defined set of profiles that it can be deployed with. For information on the resource settings defined for each component profile, see Component profiles, CPU limits, memory limits, and licensing.
Key points of component profiles:
- One replica profiles are denoted
by
n1
, and provide a single replica of each pod. The one replica profiles can be deployed on a single worker node. - Three replica profiles are denoted
by
n3
, and provide three replicas of each pod. At least three worker nodes must be available to deploy a three replica profile. If one replica pod fails, then the other two replicas continue serving requests, providing high availability. - Component profiles do not have to be the same for each component. For example, you can have a three replica deployment of the Management component that is configured with a one replica deployment of the Portal component.
- It is possible to change component profile after installation see Advanced configuration.
- Multiple Portal, Analytics, and Gateway component instances can be added with either configuration. For example, you can have a three replica Management component that is configured with two Portal services: a one replica Portal component, and a three replica Portal component.
OpenShift and Cloud Pak for Integration API Connect deployment profiles
API Connect deployment profiles define a complete API Connect deployment, specifying the component profiles for the Management, Portal, Analytics, and Gateway components to be used in that deployment.
Key points of API Connect deployment
profiles:
- The API Connect deployment profiles apply to top-level CR installations only. If you want to deploy your API Connect components individually, then select the component profiles from Component profiles, CPU limits, memory limits, and licensing. For more information about the top-level CR installation, see OpenShift: Deciding to use individual CRs, a top-level CR, or Cloud Pak for Integration.
- The API Connect deployment profile is selected by the user in the installation steps on OpenShift and Cloud Pak for Integration.
- API Connect deployment profiles provide one instance of the Portal, Analytics, and Gateway components.
- For details of each API Connect deployment profile, see: API Connect deployment profiles for OpenShift and Cloud Pak for Integration