Scaling services

You can adjust IBM® Cloud Pak for Data services by scaling the resources that they use to support high availability or to increase processing capacity. Resources can be scaled based on predefined resource configurations.

Permissions you need for this task
You must be either:
  • A cluster administrator
  • An administrator of the project where the software is installed, for example cpd-instance
When you need to complete this task
If a service supports scaling, you can scale the service at any time after you install it.

About this task

Scaling changes the capacity of services by adjusting the memory resource settings, the number of CPUs, and the number of pods that are available. Pods act as servers with that run applications or functions. Pods consume resources such as core and memory when components distribute tasks to them. Scaling the pods to medium, for example, increases the processing capacity of the application. You use the scaleConfig setting in the service custom resource (CR) to set the scaling configuration for a service. The following predefined scaling configurations are provided for services that support scaling:
  • Small (default configuration)
  • Medium
  • Large
Note: Some services might use a different default value or different predefined sizes, or have service-specific scaling methods that don't use the scaleConfig setting in the service CR. For more information, see Services that support scaling.

For services that don't support scaling, you must set the scaleConfig to NIL or None.

The relevant scale YAML files that specify the values for the different scaling configurations must be in the config-vars/scale/<arch> directory. The scaleConfig variable in the CR specifies the YAML file name without extension. Each service that supports scaling defines its own set of resources to handle various supported configurations and might include more configurations.

Services that support scaling

The following table shows the services that support scaling by using the scaleConfig setting, with more information provided for services that use a default size other than the standard default size small or support sizes other than the standard sizes small, medium, and large.

The kind and default CR name are provided for each service that uses the scaleConfig setting for scaling. You must use these values when you run the oc edit or oc patch command to change the scaling configuration of the service. If you have changed the CR name, you must use the new CR name when you run the command. The table also indicates whether a service uses a service-specific scaling method other than the scaleConfig setting.

View the Changing the scaling configuration of a service section to find service-specific scaling instructions for using the oc patch, or click on a service name link where applicable in the table to jump the instructions for that service.

Service name Uses the scaleConfig setting Uses a different scaling method More information
Cloud Pak for Data control plane
Kind: ZenService
Default CR name: lite-cr
 ✓    
Common core services
Kind: CCS
Default CR name: ccs-cr
 ✓    
Analytics Engine Powered by Apache Spark
Kind: AnalyticsEngine
Default CR name: analyticsengine-sample
 ✓    
Cognos® Analytics
Kind: CAService
Default CR name: ca-addon-cr
 ✓   Cognos Analytics supports four sizes:
  • small fixed (no scale)
  • small
  • medium
  • large
Cognos Dashboards
Kind: CdeProxyService
Default CR name: cdeproxyservice-cr
 ✓    
Data Privacy
Kind: DP
Default CR name: dp-cr
 ✓    
Data Refinery
Kind: DataRefinery
Default CR name: datarefinery-sample
 ✓  

The Data Refinery service is installed only when you install either the Watson™ Knowledge Catalog service or the Watson Studio service.
Data Virtualization
   ✓ Data Virtualization does not use the spec.scaleConfig value in the custom resource for scaling. Work with your IBM Sales representative to get an accurate sizing based on your expected workload. IBM Sales can help you estimate the total demand for Data Virtualization.

For information about the process for scaling Data Virtualization, see Scaling Data Virtualization.
DataStage®
Kind: DataStage
Default CR name: datastage
 ✓  
Db2®
   ✓

Work with your IBM Sales representative to size your cluster.

For more information about scaling Db2, see Scaling up Db2.
Db2 Big SQL
   ✓ This service is scaled by using the Db2 Big SQL instance details page in the Cloud Pak for Data web interface. Worker nodes use dynamic provisioning to allow for efficient scaling up and down of the service.

For information about scaling Db2 Big SQL, see Updating instance configuration in the Db2 Big SQL documentation.
Db2 Data Gate
Does not support scaling
      
Db2 Data Management Console
Kind: Dmcaddon
Default CR name: dmc-addon
 ✓   For information about scaling the Db2 Data Management Console service, see Scaling Db2 Data Management Console.
Db2 Event Store
Does not support scaling
      
Db2 Warehouse
   ✓

Work with your IBM Sales representative to size your cluster.

For more information about scaling Db2 Warehouse, see Scaling up Db2.
Decision Optimization
Kind: DODS
Default CR name: dods-cr
   ✓ Decision Optimization uses the scaling command to scale the service. When replicas are scaled, the CPUs and memory are also scaled.
  • 1 replica, 2.5 CPUs, 4 GB--config small
  • 2 replicas, 7 CPUs, 10 GB--config medium
  • 3 replicas, 10 CPUs, 14 GB--config large
EDB Postgres
Kind: CPDEdbService
Default CR name: cpd-edb-service
 ✓    
Execution Engine for Apache Hadoop
Does not support scaling
      
Guardium® External S-TAP®
Kind: EstapService
Default CR name: estap-cr
 ✓   Guardium External S-TAP supports the following sizes:
  • small
  • small-highmem
  • small-lowmem
  • medium
  • medium-highmem
  • medium-lowmem
  • large
  • large-highmem
  • large-lowmem
  • xlarge
  • xlarge-highmem
  • xlarge-lowmem

Specify sizes small to xlarge to manage the number of pods (horizontal scaling). Specify the amount of memory (mem, highmem, or lowmem) to manage how much memory is consumed by the pods (vertical scaling). If you do not specify sufficient memory, the pods will restart due to Out of Memory conditions.
IBM Match 360 with Watson
Kind: MasterDataManagement
Default CR name: mdm-cr
 ✓  
IBM Match 360 supports the following sizes:
  • xsmall (demonstration use only)
  • small (default)
  • medium
  • large

Work with your IBM Sales representative to get accurate sizing based on your expected workload.
Informix®
Kind: InformixService
Default CR name: informixservice-cr
 ✓    
MongoDB
Does not support scaling
      
Open Data for Industries
 ✓   Horizontal scaling is supported by OpenShift® Container Platform.
OpenPages®
   ✓
When you provision an OpenPages instance, you choose a size for the instance. The following sizes are available:
  • Extra Small (XS)
  • Small
  • Medium
  • Large

See Scaling OpenPages for detailed information about scaling the service.
Planning Analytics
Kind: PAService
Default CR name: ibm-planning-analytics-service
 ✓    
Product Master
Kind: ProductMaster
Default CR name: productmaster-cr
 ✓  

When replicas are scaled, the CPUs and memory are scaled according to the selection.

The service supports an extra large (xlarge) configuration.
Scheduling service
Does not support scaling
      
SPSS® Modeler
Kind: Spss
Default CR name: spss-sample
 ✓    
Voice Gateway
Does not support scaling
      
Watson Assistant
Kind: WatsonAssistant
Default CR name: wa
   ✓ You can specify the size of the deployment by specifying the size field in the CR. The options are small, medium, and large. The default size is medium.
Watson Discovery
Kind: WatsonDiscovery
Default CR name: wd
   ✓

You can specify size of the deployment when the CR is created by specifying the spec.shared.deploymentType as either Deployment or Production. This field is immutable field so it cannot be changed after the CR is created.
Watson Knowledge Catalog
Kind: WKC
Default CR name: wkc-cr
   ✓ To scale Watson Knowledge Catalog, first you modify the Watson Knowledge Catalog CR size setting (spec.scaleConfig) to medium or large, and then you must separately scale your Db2u instances. You must also separately scale your InfoSphere® Information Server CR.

For more information, see Scaling up Db2 for Watson Knowledge Catalog.

Legacy services always use one replica.
Watson Knowledge Studio
Kind: KnowledgeStudio
Default CR name: wks
 ✓   Watson Knowledge Studio supports small and medium configurations. The default size is small.
Watson Machine Learning
Kind: WmlBase
Default CR name: wml-cr
 ✓   Watson Machine Learning supports small and medium configurations.
Watson Machine Learning Accelerator
Kind: Wmla-add-on
Default CR name: wmla
 ✓    
Watson OpenScale
Kind: WOService
Default CR name: aiopenscale
 ✓   Watson OpenScale supports small and medium configurations. For large configurations, work with your IBM representative.
Watson Speech to Text
Kind: WatsonSpeech
Default CR name: speech-prod-cr
 ✓   Watson Speech to Text supports extra small (xsmall) configurations, which is the default size for the service.
Watson Studio
Kind: WS
Default CR name: ws-cr
 ✓    
Watson Studio Runtimes
   ✓ Watson Studio Runtimes are started with the requested resources only. You cannot scale up the started runtime. Every runtime is user and project bound.
Watson Text to Speech
Kind: WatsonSpeech
Default CR name: speech-prod-cr
 ✓   Watson Text to Speech supports extra small (xsmall) configurations, which is the default size for the service.

Changing the scaling configuration of a service

Important: Before you scale up a service, ensure that your cluster can support the additional workload. If necessary, contact your IBM Support representative.
Note: When you change the scaling configuration of a service, you also must scale the related services individually. See Service software requirements to identify the service dependencies and then scale the services as necessary.
You can scale a service by using one of the following commands:
To change the scaling configuration of a service by using the oc edit command
  1. Run the following command to edit the CR for the service:
    oc edit kind cr-name --namespace project

    Replace kind and cr-name with the values for the service. Replace project with the project where you installed Cloud Pak for Data.

  2. Add the scaleConfig variable to the service's CR. For example, to change the configuration from the default small, to medium, you would add the following setting to the spec section of the CR:

    ...
spec:
   scaleConfig: "medium"
   ...

    After the change to the CR, the scaling is updated during the next reconcile loop by an operator that is watching for the changes. For example, the configuration is changed from small to medium during the loop.

  3. For services that don't support scaling, add either NIL or None to the scaleConfig variable in the service's CR, as shown in the following example that uses NIL:
    ...
spec:
   scaleConfig: "NIL"
   ...
    The CR skips any scaling requests.
To change the scaling configuration of a service by using the oc patch command
  • Run the following command to edit the CR for the service:
    oc patch kind cr-name -n project -p '{"spec":{"scaleConfig":"size"}}' --type=merge

    Replace kind and cr-name with the values for the service.

    Replace project with the project where you installed Cloud Pak for Data.

    Replace size with the appropriate size value (small, medium, or large.

The following sections have instructions for using the oc patch command to scale specific services:

Cloud Pak for Data control plane
To scale Cloud Pak for Data control plane, run the following command:
oc patch ZenService lite-cr -n project -p '{"spec":{"scaleConfig":"size"}}' --type=merge

Change project to the project where you installed Cloud Pak for Data.

The size is small, medium, or large.

If you changed the CR name from the default lite-cr, use the new CR name in the command.

Common core services
To scale Common core services, run the following command:
oc patch CCS ccs-cr -n project -p '{"spec":{"scaleConfig":"size"}}' --type=merge

Change project to the project where you installed Cloud Pak for Data.

The size is small, medium, or large.

If you changed the CR name from the default ccs-cr, use the new CR name in the command.

Analytics Engine Powered by Apache Spark
To scale Analytics Engine Powered by Apache Spark, run the following command:
oc patch AnalyticsEngine analyticsengine-sample -n project -p '{"spec":{"scaleConfig":"size"}}' --type=merge

Change project to the project where you installed Cloud Pak for Data.

The size is small, medium, or large.

If you changed the CR name from the default analyticsengine-sample, use the new CR name in the command.

Cognos Analytics
To scale Cognos Analytics, run the following command:
oc patch CAService ca-addon-cr -n project -p '{"spec":{"scaleConfig":"size"}}' --type=merge

Change project to the project where you installed Cloud Pak for Data.

The size is small fixed (no scale), small, medium, or large.

If you changed the CR name from the default ca-addon-cr, use the new CR name in the command.

Cognos Dashboards
To scale Cognos Dashboards, run the following command:
oc patch CdeProxyService cdeproxyservice-cr -n project -p '{"spec":{"scaleConfig":"size"}}' --type=merge

Change project to the project where you installed Cloud Pak for Data.

The size is small, medium, or large.

If you changed the CR name from the default cdeproxyservice-cr, use the new CR name in the command.

Data Privacy
To scale Data Privacy, run the following command:
oc patch DP dp-cr -n project -p '{"spec":{"scaleConfig":"size"}}' --type=merge

Change project to the project where you installed Cloud Pak for Data.

The size is small, medium, or large.

If you changed the CR name from the default dp-cr, use the new CR name in the command.

Data Refinery
To scale Data Refinery, run the following command:
oc patch DataRefinery datarefinery-sample -n project -p '{"spec":{"scaleConfig":"size"}}' --type=merge

Change project to the project where you installed Cloud Pak for Data.

The size is small, medium, or large.

If you changed the CR name from the default datarefinery-sample, use the new CR name in the command.

DataStage
To scale DataStage, run the following command:
oc patch DataStage datastage -n project -p '{"spec":{"scaleConfig":"size"}}' --type=merge

Change project to the project where you installed Cloud Pak for Data.

The size is small, medium, or large.

If you changed the CR name from the default datastage, use the new CR name in the command.

Db2 Data Management Console
To scale Db2 Data Management Console, run the following command:
oc patch Dmcaddon dmc-addon -n project -p '{"spec":{"scaleConfig":"size"}}' --type=merge

Change project to the project where you installed Cloud Pak for Data.

The size is small, medium, or large.

If you changed the CR name from the default dmc-addon, use the new CR name in the command.

EDB Postgres
To scale EDB Postgres, run the following command:
oc patch CPDEdbService cpd-edb-service -n project -p '{"spec":{"scaleConfig":"size"}}' --type=merge

Change project to the project where you installed Cloud Pak for Data.

The size is small, medium, or large.

If you changed the CR name from the default cpd-edb-service, use the new CR name in the command.

Guardium External S-TAP
To scale Guardium External S-TAP, run the following command:
oc patch EstapService estap-cr -n project -p '{"spec":{"scaleConfig":"size"}}' --type=merge

Change project to the project where you installed Cloud Pak for Data.

The size is one of the following values:
  • small
  • small-highmem
  • small-lowmem
  • medium
  • medium-highmem
  • medium-lowmem
  • large
  • large-highmem
  • large-lowmem
  • xlarge
  • xlarge-highmem
  • xlarge-lowmem

If you changed the CR name from the default estap-cr, use the new CR name in the command.

IBM Match 360 with Watson
To scale IBM Match 360 with Watson, run the following command:
oc patch MasterDataManagement mdm-cr -n project -p '{"spec":{"scaleConfig":"size"}}' --type=merge

Change project to the project where you installed Cloud Pak for Data.

The size is xsmall (demonstration use), small, medium, or large.

If you changed the CR name from the default mdm-cr, use the new CR name in the command.

Informix
To scale Informix, run the following command:
oc patch InformixService informixservice-cr -n project -p '{"spec":{"scaleConfig":"size"}}' --type=merge

Change project to the project where you installed Cloud Pak for Data.

The size is small, medium, or large.

If you changed the CR name from the default informixservice-cr, use the new CR name in the command.

Planning Analytics
To scale Planning Analytics, run the following command:
oc patch PAService ibm-planning-analytics-service -n project -p '{"spec":{"scaleConfig":"size"}}' --type=merge

Change project to the project where you installed Cloud Pak for Data.

The size is small, medium, or large.

If you changed the CR name from the default ibm-planning-analytics-service, use the new CR name in the command.

Product Master
To scale Product Master, run the following command:
oc patch ProductMaster productmaster-cr -n project -p '{"spec":{"scaleConfig":"size"}}' --type=merge

Change project to the project where you installed Cloud Pak for Data.

The size is small, medium, large, or xlarge.

If you changed the CR name from the default productmaster-cr, use the new CR name in the command.

SPSS Modeler
To scale SPSS Modeler, run the following command:
oc patch Spss spss-sample -n project -p '{"spec":{"scaleConfig":"size"}}' --type=merge

Change project to the project where you installed Cloud Pak for Data.

The size is small, medium, or large.

If you changed the CR name from the default spss-sample, use the new CR name in the command.

Watson Knowledge Studio
To scale Watson Knowledge Studio, run the following command:
oc patch KnowledgeStudio wks -n project -p '{"spec":{"scaleConfig":"size"}}' --type=merge

Change project to the project where you installed Cloud Pak for Data.

The size is either small or medium.

If you changed the CR name from the default wks, use the new CR name in the command.

Watson Machine Learning
To scale Watson Machine Learning, run the following command:
oc patch WmlBase wml-cr -n project -p '{"spec":{"scaleConfig":"size"}}' --type=merge

Change project to the project where you installed Cloud Pak for Data.

The size is either small or medium.

If you changed the CR name from the default wml-cr, use the new CR name in the command.

Watson Machine Learning Accelerator
To scale Watson Machine Learning Accelerator, run the following command:
oc patch Wmla-add-on wmla -n project -p '{"spec":{"scaleConfig":"size"}}' --type=merge

Change project to the project where you installed Cloud Pak for Data.

The size is small, medium, or large.

If you changed the CR name from the default wmla, use the new CR name in the command.

Watson OpenScale
To scale Watson OpenScale, run the following command:
oc patch WOService aiopenscale -n project -p '{"spec":{"scaleConfig":"size"}}' --type=merge

Change project to the project where you installed Cloud Pak for Data.

The size is either small or medium.

If you changed the CR name from the default aiopenscale, use the new CR name in the command.

Watson Text to Speech
To scale Watson Text to Speech, run the following command:
oc patch WatsonSpeech speech-prod-cr -n project -p '{"spec":{"scaleConfig":"size"}}' --type=merge

Change project to the project where you installed Cloud Pak for Data.

The size is xsmall, small, medium, or large.

If you changed the CR name from the default speech-prod-cr, use the new CR name in the command.

Watson Studio
To scale Watson Studio, run the following command:
oc patch WS ws-cr -n project -p '{"spec":{"scaleConfig":"size"}}' --type=merge

Change project to the project where you installed Cloud Pak for Data.

The size is small, medium, or large.

If you changed the CR name from the default ws-cr, use the new CR name in the command.