Changing Power settings

Important: IBM Cloud Pak® for Data Version 4.8 will reach end of support (EOS) on 31 July, 2025. For more information, see the Discontinuance of service announcement for IBM Cloud Pak for Data Version 4.X.

Upgrade to IBM Software Hub Version 5.1 before IBM Cloud Pak for Data Version 4.8 reaches end of support. For more information, see Upgrading from IBM Cloud Pak for Data Version 4.8 to IBM Software Hub Version 5.1.

On Power® Systems, you must change the simultaneous multithreading (SMT) settings for Kernel-based Virtual Machine (KVM) capable systems and big core PowerVM® capable systems.

Installation phase

Setting up a client workstation

Setting up a cluster

Collecting required information

Preparing to run installs in a restricted network

Preparing to run installs from a private container registry

Preparing the cluster for Cloud Pak for Data

Preparing to install an instance of Cloud Pak for Data

Installing an instance of Cloud Pak for Data

Who needs to complete this task?

Cluster administrator A cluster administrator must complete this task.

When do you need to complete this task?

One-time setup Complete this task if you plan to install Cloud Pak for Data on the following environments:

Kernel-based Virtual Machine (KVM) capable systems: LC922, IC922, and AC922
Big core PowerVM capable systems: L922, E950, E980, S922, S1022, S1024, E1050, and E1080

Procedure

Complete the following steps to change the simultaneous multithreading (SMT) settings.

The SMT settings depend on whether you plan to install the Db2® Warehouse service. If you plan to install Db2 Warehouse, consider whether you plan to install the service on dedicated nodes.

Label all small core KVM capable worker nodes by using one of the following options:
- If the worker node will not run Db2 Warehouse workloads, use SMT=2, as shown in the following example:
```
oc label node <node> SMT=2 --overwrite
```
- If the worker node will run Db2 Warehouse workloads, use SMT=4, as shown in the following example:
```
oc label node <node> SMT=4 --overwrite
```
Label all big core PowerVM capable worker nodes by using one of the following options:
- If the worker node will not run Db2 Warehouse workloads, use SMT=4, as shown in the following example:
```
oc label node <node> SMT=4 --overwrite
```
- If the worker node will run Db2 Warehouse workloads, use SMT=8, as shown in the following example:
```
oc label node <node> SMT=8 --overwrite 
```

Create a YAML file, smt.yaml, with the following content:

apiVersion: machineconfiguration.openshift.io/v1
kind: MachineConfig
metadata:
  labels:
    machineconfiguration.openshift.io/role: worker
  name: 99-worker-smt
spec:
  config:
    ignition:
      version: 3.1.0
    storage:
      files:
      - contents:
          source: data:text/plain;charset=utf-8;base64,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
          verification: {}
        filesystem: root
        mode: 0755
        overwrite: true
        path: /usr/local/bin/powersmt
    systemd:
      units:
        - name: smt.service
          enabled: true
          contents: |
            [Unit]
            Description=Set SMT
            After=network-online.target
            Before= crio.service
            [Service]
            Type=oneshot
            RemainAfterExit=yes
            ExecStart=/usr/local/bin/powersmt
            [Install]
            WantedBy=multi-user.target

Ensure that the control plane nodes on the cluster are Ready:
```
oc get nodes
```
Apply the settings from the smt.yaml file:
```
oc create -f smt.yaml
```
When you run this command, the cluster performs a rolling restart of all of the worker nodes to update the slub_max_order kernel argument and set the specified SMT level.

If a worker node is not labeled, the node uses the default SMT level.

If you need to change the SMT level on a specific node, label the worker node with the correct SMT level and manually restart the node.