Skip to content

IBM Cloud Private environment

If you want setting up the environment on your own you will need at least seven VMs:

  • One ICP cluster with at the bare minimum of three VMs: one master-proxy node, two worker nodes. This is for the run time.
  • For the machine learning part, we propose to use a dedicated IBM Cloud Pak for data, with four nodes: ne master-proxy node, three worker nodes.
  • One VM for DB2 community edition and liberty server

In production, you will use a high availability clusters with 3 masters, 3 managements, 3 proxy, at least 4 worker nodes per ICP clusters.

Installing IBM Cloud Private

You can study this website and reuse the Terraform file to configure your environment.

Installing IBM Cloud Pack for Data

We do not plan to add yet another step by step instruction for installing ICP4D. The product installation instructions are at https://docs-icpdata.mybluemix.net/install/com.ibm.icpdata.doc/zen/install/standovu.html

and to prepare VMs see this note

The ICP run time clustering

The following diagram illustrates the ICP cluster topology with some of the major ICP components:

The dashed lines highlight the deployment concept of k8s. The Db2 warehouse is using external Glusterfs cluster for persisting data via the persistent volumes and persistent volume claim.

The spark cluster, master, spawner... are deployments inside ICP and installed via DSX Local.