Search

Chapter 1. Introduction to Red Hat Ansible Lightspeed

download PDF

Learn about Red Hat Ansible Lightspeed with IBM watsonx Code Assistant, its benefits, key features, process, and data gathered to train the IBM watsonx Code Assistant models.

1.1. About Red Hat Ansible Lightspeed

Red Hat Ansible Lightspeed with IBM watsonx Code Assistant is a generative AI service that helps automation teams create, adopt, and maintain Ansible content more efficiently. It uses natural language prompts to generate code recommendations for automation tasks based on Ansible best practices.

Red Hat Ansible Lightspeed is the cloud service that enables integration of generative AI into Ansible Automation Platform. This document specifically describes the integration of Red Hat Ansible Lightspeed with IBM watsonx Code Assistant.

Red Hat Ansible Lightspeed uses IBM watsonx Code Assistant models trained on subject matter expertise across the Ansible ecosystem, which includes Galaxy, GitHub, and Ansible certified and validated content. For ease of use, Red Hat Ansible Lightspeed is integrated with your existing Ansible developer workflows. For example, you can use your existing Git repositories (both public and private) to train your IBM watsonx Code Assistant models. You can also access Lightspeed content suggestions in VS Code through the Ansible VS code extension.

1.1.1. Accessing Red Hat Ansible Lightspeed with IBM watsonx Code Assistant

To use Red Hat Ansible Lightspeed with IBM watsonx Code Assistant, your organization must have:

  • A trial or paid subscription to Ansible Automation Platform
  • A trial or paid subscription to IBM watsonx Code Assistant

1.1.2. Benefits of using Red Hat Ansible Lightspeed

Red Hat Ansible Lightspeed with IBM watsonx Code Assistant offers the following benefits:

  • Reduces the onboarding learning period for Ansible developers

    With just a basic understanding of YAML syntax, Ansible developers can use natural language prompts in English language to describe the automation goal. Red Hat Ansible Lightspeed then offers Ansible code recommendations to help achieve the automation goal more efficiently. This combination of content and best practice suggestions reduces the learning curve and offers a smoother onboarding experience for new Ansible users.

    For example, to get a multitask code recommendation, you can enter the prompt Install postgresql-server & run postgresql-setup command. The Ansible Lightspeed service reads the text, interacts with IBM watsonx Code Assistant, and generates code recommendations to automate a multitask that installs a PostgreSQL server and sets up a PostgreSQL database. You can then view and accept the code recommendations to create tasks in an Ansible YAML file.

  • Increases productivity with quality content creation

    Red Hat Ansible Lightspeed offers automation code recommendations that adhere to Ansible best practices, and IBM watsonx Code Assistant provides model fine-tuning features to improve the accuracy of suggested content based on your organization’s existing Ansible content. Therefore, the AI-generated code recommendations are more accurate, more reliable, and integrated with your existing automation development workflows.

  • Extends trust with AI-generated code recommendations

    The AI-generated code recommendations enable you to extend trust, with an automation code base that adheres to accepted Ansible best practices and significant data safeguards.

1.2. Key features of Red Hat Ansible Lightspeed

Red Hat Ansible Lightspeed offers the following key features:

  • Ansible-specific IBM watsonx Code Assistant models

    Red Hat Ansible Lightspeed with IBM watsonx Code Assistant uses Ansible-specific IBM watsonx Granite models unique to your organization, which are provided, managed, and maintained by IBM.

  • Model customization

    Organization administrators can now create and use fine-tuned, custom models that are trained on your organization’s existing Ansible content. With this capability, you can tune the models to your organization’s automation patterns and improve the code recommendation experience.

    You can configure multiple custom models for your organization. For example, you can create a custom model for your corporate IT automation team and a different one for your engineering team’s infrastructure. You can also configure a custom model to make it available for all Ansible users or select Ansible users in your organization.

  • Red Hat Ansible Lightspeed cloud service and on-premise deployments

    Red Hat Ansible Lightspeed is available both as a cloud service and as an on-premise deployment. Red Hat Ansible Lightspeed on-premise deployments provide the Red Hat Ansible Automation Platform customers more control over their data and supports compliance with enterprise security policies. For example, organizations in sensitive industries with data privacy or air-gapped requirements can use on-premise deployments of both Red Hat Ansible Lightspeed and IBM watsonx Code Assistant for Red Hat Ansible Lightspeed on Cloud Pak for Data. Red Hat Ansible Lightspeed on-premise deployments are supported on Red Hat Ansible Automation Platform version 2.4.

  • Playbook and task generation

    This includes the following capabilities:

    • Playbook generation and explanations

      Using the Ansible VS Code extension, you can create Ansible playbooks using a natural language interface in English. Red Hat Ansible Lightspeed with IBM watsonx Code Assistant reads the natural language prompts and generates an entire playbook recommendation based on your intent. You can also view the explanations for new or existing playbooks. The playbook explanations describe what the playbook or task within the playbook does and contextualize its impact.

    • Single and multitask generation

      Using natural language prompts, you can generate single task or multiple task recommendations for Ansible task files and playbooks. To request multitask code recommendations, you can enter a sequence of natural language task prompts in a YAML file comment separated by ampersand (&) symbols.

      Currently, Red Hat Ansible Lightspeed supports user prompts in English language only. However, there could be instances where the training data that was used to train the IBM watsonx Code Assistant models included non-English language. In such scenarios, the model can generate code recommendations for prompts made in the same non-English language, but the generated code recommendations might or might not be accurate.

  • Content source matching

    For each generated code recommendation, Red Hat Ansible Lightspeed lists content source matches, including details such as potential source, content author, and relevant licenses. You can use this data to gain insight into potential training data sources used to generate the code recommendations.

  • Post-processing capabilities

    Red Hat Ansible Lightspeed offers post-processing capabilities that augment IBM watsonx Code Assistant and improve the quality and accuracy of code recommendations.

  • Content maintenance and modernization

    The Ansible code bot scans existing content collections, roles, and playbooks through Git repositories, and proactively creates pull requests whenever best practices or quality improvement recommendations are available. The bot automatically submits pull requests to the repository, which proactively alerts the repository owner to a recommended change to their content.

  • Telemetry data collection on the Admin dashboard

    Red Hat Ansible Lightspeed now collects Admin dashboard telemetry data that provides insight into how your organization users are using the Ansible Lightspeed service, and displays the metrics on the Admin dashboard. If you no longer want to collect and manage the Admin dashboard telemetry, you can disable it for your organization.

1.3. Using Red Hat Ansible Lightspeed with IBM watsonx Code Assistant

1.3.1. Prerequisites

To use Red Hat Ansible Lightspeed cloud service, ensure that you have the following components:

  • Trial or paid subscription to Ansible Automation Platform
  • Trial or paid subscription to IBM watsonx Code Assistant
  • VS Code version 1.70.1 or later
  • The Ansible extension for VS Code version 2.8 or later

To use an on-premise deployment of Red Hat Ansible Lightspeed, your organization must have the following subscriptions:

  • A trial or paid subscription to Red Hat Ansible Automation Platform
  • An installation of IBM watsonx Code Assistant for Red Hat Ansible Lightspeed on Cloud Pak for Data

1.3.2. Connectivity requirements

To generate code recommendations, the Ansible Lightspeed service in Visual Studio (VS) Code editor requires access to the following outbound domains:

The outbound connections are encrypted on TCP protocol port 443.

1.4. Data gathered to train the IBM watsonx Code Assistant models

1.4.1. Models

Red Hat Ansible Lightspeed with IBM watsonx Code Assistant uses Ansible-specific IBM watsonx Granite models unique to your organization. These models are provided, managed, and maintained by IBM.

1.4.2. Data sources

IBM watsonx Code Assistant models are trained on Ansible content from Ansible Galaxy, data from public Git repositories, and Red Hat Ansible subject matter expert examples.

If you publish content to Ansible Galaxy and want to restrict your Ansible Galaxy content from being used to train the models, you can opt out of sharing your Ansible Galaxy data in the Ansible Galaxy namespace configuration.

1.4.3. Data telemetry

Red Hat Ansible Lightspeed collects the following telemetry data by default:

  • Operational telemetry data

    This is the data that is required to operate and troubleshoot the Ansible Lightspeed service. For more information, refer the Enterprise Agreement. You cannot disable the collection of operational telemetry data.

  • Admin dashboard telemetry data

    This is the data that provides insight into how your organization users are using the Ansible Lightspeed service, and the metrics are displayed on the Admin dashboard. You can also disable the Admin dashboard telemetry if you no longer want to collect and monitor the telemetry data. For more information about Admin dashboard telemetry, see Viewing and managing Admin dashboard telemetry.

1.4.4. Telemetry data collection notice for the Admin dashboard

In connection with your use of this Red Hat offering, Red Hat may collect telemetry data about your use of the software. This data allows Red Hat to monitor the software and to improve Red Hat offerings and support, including identifying, troubleshooting, and responding to issues that impact users. The data may also be used to enable you to track your entitlements to Red Hat subscriptions and take advantage of future Red Hat purchasing programs. It may also allow Red Hat to assist you in implementing upgrades to minimize service impact. The data may be shared internally within Red Hat to improve the user experience. If you are evaluating Red Hat software, the data will help Red Hat determine if you need assistance.

1.4.4.1. What information does Red Hat collect?

Tools within the software monitor various metrics and this information is transmitted to Red Hat. The following metrics are monitored:

  • Organization you are logged into (Organization ID, account number)
  • Large language model (or models) that you are connected to
  • Prompts and content suggestions, including accept or reject of the content suggestions
  • User sentiment feedback

1.4.4.2. Personal Data

Red Hat does not intend to collect personal information. If Red Hat discovers that personal information has been inadvertently received, Red Hat will delete such information. To the extent that any telemetry data constitutes personal data, refer the Red Hat Privacy Statement for more information about Red Hat’s privacy practices.

1.4.4.2.1. Retention

Red Hat retains and stores telemetry data only for as long as it’s needed for the purposes described above or as otherwise required or permitted by law.

1.4.4.2.2. Data Security

Red Hat employs technical and organizational measures designed to protect the telemetry data. Data stored in the Red Hat cloud is being protected, where possible, through encryption. Data is also segmented, and therefore is not accessible across organizations.

1.4.4.2.3. Data Sharing

Red Hat may share telemetry data with its business partners in an aggregated form that does not identify customers to help the partners better understand their markets and their customer’s use of Red Hat offerings or ensure the successful integration of products jointly supported by those partners.

1.4.4.2.4. Third Party Service Providers

Red Hat may engage certain service providers to assist in the collection and storage of the telemetry data.

1.4.4.2.5. User Control/ Enabling and Disabling Admin Dashboard Telemetry Collection

You cannot disable collection of operational telemetry data. Operational telemetry data includes only data that is necessary to operate and troubleshoot the service. However, you can disable the collection of Admin Dashboard telemetry data. For more information, see Disabling the Admin dashboard telemetry.

Red Hat logoGithubRedditYoutubeTwitter

Learn

Try, buy, & sell

Communities

About Red Hat Documentation

We help Red Hat users innovate and achieve their goals with our products and services with content they can trust.

Making open source more inclusive

Red Hat is committed to replacing problematic language in our code, documentation, and web properties. For more details, see the Red Hat Blog.

About Red Hat

We deliver hardened solutions that make it easier for enterprises to work across platforms and environments, from the core datacenter to the network edge.

© 2024 Red Hat, Inc.