Skip to content

IBM Turbonomic and CP4Data: Evaluation Edition Details and Setup

Eva Tuczai edited this page Dec 14, 2023 · 10 revisions

IBM Turbonomic Visibility and Optimization For OpenShift and Cloud Pak for Data

Thank you for being an IBM Cloud Pak customer and taking advantage of a limited time free evaluation of IBM Turbonomic. This document will describe what features you can experience in an Evaluation. The Deployment and Configuration section will walk you through setup steps for the Turbonomic Server and the Kubeturbo probe, both of which will deploy into your OpenShift Cluster.

Table of Contents

Overview

IBM Cloud Pak for Datais a modular set of integrated software components for data analysis, organization and management, that is designed to the leverage modern digital platform of Red Hat OpenShift Container Platform

IBM Turbonomic provides Kubernetes Optimization for both the OpenShift Platform and the microservices applications that run there.

Why Turbonomic: Optimizing Apps and Platforms

IBM Turbonomic is the premier solution for Application Resource Management (ARM) of cloud and virtual environments. With Turbonomic, businesses can receive actionable insights using AI to improve performance, stay compliant with business policies, and eliminate waste. Turbonomic will assist businesses in efficiently managing resources such as containers, Virtual Machines (VMs), servers, storage, networks, and databases to assure performance and minimize costs.

Turbonomic is now available for integration with Cloud Pak for Data for monitoring the OpenShift clusters and helping customers achieve optimal performance in dynamic multi-cloud or containerized environments. Cloud Pak for Data runs on the Red Hat OpenShift container platform and uses the Kubernetes orchestrator’s capabilities to configure applications running in containers, making it possible to deploy and operate them from any certified Cloud. Turbonomic installs onto a Red Hat OpenShift cluster in a single namespace deployment. Turbonomic automatically determines the right resource allocation actions to make recommendations around how to scale, provision, deprovision and optimize resources for performance, cost, and availability. This allows Cloud Pak for Data administrators to reap the benefits of elasticity, resiliency, and speed to market in dynamic multi-cloud environments.

Why Turbo for CP4D: Better Together

Kubernetes is self-healing but not self-optimizing. While Kubernetes helps customers reap the benefits of application elasticity, it can be hard to control platform complexity to effectively manage fluctuating workload demand. The ephemeral nature of containerized workload and infrastructure creates resourcing challenges to assure performance without over-provisioning. This is where Turbonomic comes in, simplifying the process by automatically determining the right resourcing actions at the right time, whether it is rightsizing or proactively redistributing pods or intelligently scaling workloads and nodes. Turbonomic also provides robust observability of key resource and performance data that is also behind the analysis.

To read more about how to use Turbonomic to spot resourcing issues in CP4D, you can also refer to these articles:

Evaluation Edition for CP4Data

What do I get with the Evaluation Edition of IBM Turbonomi? You will get access to most of the features of Turbonomic but will be restricted to not be able to execute actions, and this key will expire on Dec 30, 2024. The following table provides a high-level comparison of features in the free vs licensed version.

Feature Function Evaluation Edition Premium (Upgrade)
Monitoring of CP4D containerized application components and OCP Cluster Visibility and Observability YES YES
Analysis and actions to manage tradeoffs of performance and efficiency Mitigate performance risk and identify best efficiency opportunities. Reduce toil to optimize YES YES
Simulate what-if scenarios Planning for change, capacity planning and seeing the results of taking actions in a simulation YES YES
Multiple Target Types including APM Tools Provide full stack visibility, context and optimization across all IT stacks. Limited (OCP, AWS, Azure, Google, vCenter, Instana) YES-All
Executable Actions Automate optimization Actions in RECOMMEND only YES-All
Target your environment Optimization across Dev, Prod, and all environments Limited: 5000 VMs and/or OCP Nodes As per licensed capacity
Policies Customize the analysis and action execution View but cannot edit Policies YES-All
Reporting Extend visibility into showback reports, summary of value obtained, detailed data Limited - single report YES

This evaluation mode is upgradable with a software license purchase. Simply apply the new license key, and Premium features will automatically be available.

Deployment and Configuration: Server

The preferred method is to leverage the OpenShift Operator Hub of Certified Operators.

References

This document will make use of the following on-line references:

Prerequisites and Checklist

  • OpenShift Cluster:
    • V4.11 and higher
    • available capacity of 2 nodes with 32 Gb Mem x 4 CPUs each
    • X86/AMD running Linux
  • Storage Class configured with Retain reclaim policy. Block storage preferred.
  • Access to OpenShift Operator Hub (OCP Operator Hub) to use Turbonomic Certified operators.
  • Access to pull container images from IBM’s public repo: icr.io/cpopen/turbonomic
  • Provide XL Custom Resource to deploy a Turbonomic Server instance. CR yaml is located here
  • Obtain an evaluation license from Turbonomic. Refer to Obtain License section for more details.

Server Installation Steps

The method used will deploy a minimal Server with a containerized database, and will also provide the ability to add AWS, Azure, Google and vCenter targets provided they stay within the evaluation license limits of 5000 Nodes / Virtual Machines.

  1. Create a namespace. Examples will use turbonomic
  2. In the OCP Operator Hub, search under “Turbonomic” and select “Turbononic Platform Certified Operator”.
  3. Install the operator into the namespace you created.
  • You can choose to auto-update
  1. Before you deploy the server, we have to set scc context at a group level otherwise the Grafana reporting will not work (an issue with Grafana and inheriting scc). In a terminal session, log into your OCP cluster using oc login and run oc adm policy add-scc-to-group anyuid system:serviceaccounts:turbonomic. If your namespace / project is different from turbonomic, please update the example.
  2. Using the sample XL Custom Resource yaml provided, edit the following parameters that are all under global:
  • fsGroup – Obtain the YOUR-NS-UID from the Namespace's sa.scc.uid-range using the numerator. For example, 1000020000
  • under postgresql - Provide the same YOUR-NS-UID value for two runAsUser values.
  • storageClassName - Provide a storage class name that provides the capabilities required.
  • tag - Supply a value that is equal to the Latest Version identified from here. For example, 8.10.4
  • Note: when you want to update the Turbo Server, you will modify this value to reflect the latest version.
  1. Apply the XL CR. Validate the running environment with the documentation provided here
  2. Identify the URL to launch Turbo UI by going to ROUTES in the OCP Console for your Turbonomic namespace. You will see a Route called nginx with a Location naming convention of https://nginx-NAMESPACE.apps.YOURCLUSTERDOMAIN. Click on this Location link.
  3. For first time log in, you will be prompted to create a password for the default administrator user.
  • Store this password in a safe place. You cannot reset this password. If you lose the password you must uninstall and reinstall the Server
  1. Next in the UI you will be prompted to supply the license key you obtained from Turbonomic.

Now proceed to setting up the Kubeturbo client.

Deployment and Configuration: Kubeturbo Client

Set up Kubeturbo first before any other Target. The method used will deploy Kubeturbo probe into the OpenShift Cluster where CP4Data is running. To leverage Turbonomic, deploy the Kubeturbo mediation probe into the OCP cluster where Cloud Pak for Data is running.

References

This document will make use of the following on-line references:

Prerequisites and Checklist

  • Running Turbonomic Server with the license key applied

  • You have a local username and password from the Turbonomic server that can be used for registering the Kubeturbo probe. This user requires Administrator or Site Administrator role. We will use the administrator created when you installed the Turbo Server

  • OpenShift Cluster:

    • V4.9 and higher
    • Supported archs of X86/AMD, ARM, PPC, OS390, running Linux
  • Access to OpenShift Operator Hub (OCP Operator Hub) to use Turbonomic Certified operators.

  • Access to pull container images from IBM’s public repo: icr.io/cpopen/turbonomic

  • Provide Kubeturbo Custom Resource to deploy a Kubeturbo instance. CR yaml is located here.

Client Installation Steps

  1. Create a namespace. Examples will use kubeturbo
  2. In the OCP Operator Hub, search under “Turbonomic” and select “Kubeturbo Certified Operator.”
  3. Install the operator into the namespace you created.
  • Select stable channel
  • You can choose to auto-update. This will also auto-update the Kubeturbo client when the operator updates
  1. In the namespace you created, create a Secret with the following properties:
  • Key/value type
  • Secret name: turbonomic-credentials
  • Key 1
    • username: administrator
  • Key 2
    • password: YOURPASSWORD
  • See this wiki article for more details
  1. Using the sample Kubeturbo Custom Resource yaml provided, edit the following parameter under serverMeta:
  • turboServer – Provide the same nginx Route Location URL to access the Turbo UI (note remove any “/” at the end of the URL)
  1. Apply the Kubeturbo CR. There will be a pod starting with Kubeturbo-cp4d-evaluation being created.
  2. In the Turbonomic UI, you can go to Settings -> Target Configuration and you should see a target called Kubernetes-CP4D-Cluster registered.

Note if you want to add another OpenShift cluster, you can reuse the same Kubeturbo Custom Resource in another cluster, but you must change the targetName parameter. Every cluster added needs to have a unique name.

Wait 24 hours for the system to collect data, and start experiencing how Turbonomic can turn data into recommendations that optimize the tradeoffs of assuring performance while being as efficient as possible.

Report Configuration

This evaluation edition will entitle you to leverage a Container and Pod usage report that you can filter by application labels. We will set up reporting and import the configuration file to set up the report for you.

  1. Designate a Turbo user as a Report Editor
  • In the Turbonomic UI, navigate in the left hand nav bar to Settings -> User Management
  • Click the administrator default user account.
  • Under Options, choose DESIGNATE AS REPORT EDITOR
  • Click Save User to save these settings
  1. Import the report config
  • In the Turbonomic UI, navigate in the left hand nav bar to Reports (Legacy). This will launch a new browser tab with the Grafana dashboard
  • Navigate to Dashboards page. On this page, to the right of the search bar, click New drop down button and select New Folder to create a folder where you will place the report configuration JSON format file. Name this folder anything you like. Click Create.

Note if you do not see the New drop down button, close the window, log out of Turbonomic, log back in to Turbonomic and relaunch Reports.

  • Download this JSON file
  • In this new folder, click on the New drop down button and select Import
  • Drag the downloaded JSON file to the clip board. The Name that will be populated by default will be Kubernetes Container Usage - with tag selection. Select Turbo Timescale as the Datasource. Click Import.
  • Note you may encounter an error if there is no data yet. These errors will go away after 24 hours of data collection.

Obtain License

To obtain a license key for the Evaluation Edition of the IBM Turbonomic server, please send an email to the Product Manager:

[email protected]

Subject: Turbo Evaluation Edition License Key Request

Please provide:

  • Your Name
  • Role / Job Title
  • Work email address
  • IBM Customer Number (aka IBM Account CMR Number)

An email with the license key file will be sent within 1 business day.

If you are interested in learning how to upgrade the license to a fully functional version, please email the Product Manager [email protected] with the Subject: Turbonomic Evaluation Upgrade.

Updating the Product

Turbonomic releases an update every two weeks, and this evaluation edition entitles you to take advantage of latest releases. To determine the latest current version, go to this link https://www.ibm.com/docs/en/tarm/latest?topic=documentation-all-turbonomic-versions and identify the version stated under “Latest Version”. These steps will show you how to update the Server and Kubeturbo client.

Updating Turbonomic Server

Do this first.

  1. In OpenShift Console go to Installed Operators, find "Turbonomic Platform Operator" and select.
  2. Go to Subscription to see if a new version is available. If you are on autoupdate, skip to next step. If on manual, then follow the prompts to upgrade the operator
  3. Go to Turbonomic Platform Operator and click on cp4d-evaluation which is your instance of a Turbonomic Server.
  4. Go to YAML and then modify the tag parameter which is under spec:global: (approximately line 120). Change the value to the latest product version, such as 8.10.5.
  5. Wait for all pods to be Running and Ready (1/1). Proceed to update Kubeturbo

Updating Kubeturbo

  1. In OpenShift Console go to Installed Operators, find "Kubeturbo Operator" and select.
  2. Go to Subscription to see if a new version is available. If you are on autoupdate, skip to next step. If on manual, then follow the prompts to upgrade the operator
  3. Go to Kubeturbo Operator and click on kubeturbo-cp4d-evaluation which is your instance of a Kubeturbo client on this OCP cluster.
  4. Go to YAML and then modify the tag parameter which is under spec:global: (approximately line 120). Change the value to the latest product version, such as 8.10.5.

Note this version must match the Turbo Server version Note if you do not see a tag parameter, then your Kubeturbo probe will automatically update to the latest version when you update the Kubeturbo Operator.

  1. Wait for all pods to be Running and Ready (1/1).

Contact and Support

As a user of this free evaluation edition, you can leverage the following resources:

Questions on installation and configuration, please refer to

  1. IBM Turbonomic Community
  2. Kubeturbo troubleshooting here
  3. Knowledge base Articles here

Questions on how to use Turbonomic, please refer to

  1. IBM Turbonomic Community
  2. Kubeturbo Use Cases here
  3. Kubeturbo action details here

Turbonomic releases an update every two weeks. For more information on how to update:

  1. Release Notes from the latest edition available from the online documentation.
  2. Updating Turbonomic Server go here
  3. Updating Kubeturbo Client manually (not on autoupdate) go here

Kubeturbo

Introduction
  1. What's new
  2. Supported Platforms
Kubeturbo Use Cases
  1. Overview
  2. Getting Started
  3. Full Stack Management
  4. Optimized Vertical Scaling
  5. Effective Cluster Management
  6. Intelligent SLO Scaling
  7. Proactive Rescheduling
  8. Better Cost Management
  9. GitOps Integration
  10. Observability and Reporting
Kubeturbo Deployment
  1. Deployment Options Overview
  2. Prerequisites
  3. Turbonomic Server Credentials
  4. Deployment by Helm Chart
    a. Updating Kubeturbo image
  5. Deployment by Yaml
    a. Updating Kubeturbo image
  6. Deployment by Operator
    a. Updating Kubeturbo image
  7. Deployment by Red Hat OpenShift OperatorHub
    a. Updating Kubeturbo image
Kubeturbo Config Details and Custom Configurations
  1. Turbonomic Server Credentials
  2. Working with a Private Repo
  3. Node Roles: Control Suspend and HA Placement
  4. CPU Frequency Getter Job Details
  5. Logging
  6. Actions and Special Cases
Actions and how to leverage them
  1. Overview
  2. Resizing or Vertical Scaling of Containerized Workloads
    a. DeploymentConfigs with manual triggers in OpenShift Environments
  3. Node Provision and Suspend (Cluster Scaling)
  4. SLO Horizontal Scaling
  5. Turbonomic Pod Moves (continuous rescheduling)
  6. Pod move action technical details
    a. Red Hat Openshift Environments
    b. Pods with PVs
IBM Cloud Pak for Data & Kubeturbo:Evaluation Edition
Troubleshooting
  1. Startup and Connectivity Issues
  2. KubeTurbo Health Notification
  3. Logging: kubeturbo log collection and configuration options
  4. Startup or Validation Issues
  5. Stitching Issues
  6. Data Collection Issues
  7. Collect data for investigating Kubernetes deployment issue
  8. Changes to Cluster Role Names and Cluster Role Binding Names
Kubeturbo and Server version mapping
  1. Turbonomic - Kubeturbo version mappings
Clone this wiki locally