The CIO Organization used IBM® watsonx Assistant™ for Z, a generative artificial intelligence (gen AI) assistant, to help bridge the skills gap to manage the IBM Z environment. The gen AI assistant is augmented with curated content which is grounded in IBM Z-specific knowledge and best practices.
The organization collaborated with the IBM product team to understand and explore how to use gen AI to transform the way enterprise users, with varying experience levels, interact with IBM Z. Through a series of brainstorming sessions, both teams prioritized 3 of 35 use cases for initial adoption and deployed watsonx Assistant for Z on-premises on an existing Red Hat® OpenShift® cluster.
As part of the partnership with the product team, the CIO Organization provided feedback to the watsonx Assistant for Z team to enhance product features and capabilities based on implementation findings and results.
The details of the watsonx Assistant for Z implementation are best showcased in these initial 3 use cases (more to come as new use cases are rollout this year):
- IBM Z questions and answers (Q&A) support for operations staff
- Before: IBM Z operations staff asked SMEs questions with answers that can be found in product-related documentation.
- After: watsonx Assistant for Z provides digestible and actionable answers with links to learn more, allowing the SMEs to focus on more strategic work.
- Incident analysis
- Before: IBM Z support team manually searched through past incidents for similar issues and their resolutions.
- After: watsonx Assistant for Z ingests the incident from a third-party workflow automation platform, analyzes the problem and presents suggested courses of action based on product documentation and previous incident analysis on similar systems.
- Patching IBM Db2 subsystems
- Before: IBM Z operations staff manually entered multiple pieces of information into a form to open a change request, receive multiple approvals, trigger automation, back-out if failed and close change request.
- After: IBM Z operations staff tell watsonx Assistant for Z that a system needs a patch. It then retrieves details from a third-party workflow automation platform and looks up relevant automation in Ansible®. After the staff specifies when to apply the patching, the gen AI assistant opens a standard change request, sends relevant details to trigger automation, prompts additional automation on failure, and automatically closes the change request with the correct status. Lastly, the staff confirms the patching was applied.
There are 2 main reasons why the CIO Organization chose watsonx Assistant for Z:
- IBM watsonx Assistant for Z not only puts decades of mainframe experience and best practices at the team's fingertips, providing answers to their queries, but also offer authorized users an option to trigger automation to perform tasks.
- The model used by IBM watsonx Assistant for Z addresses a common cause of potential hallucinations by leveraging Retrieval Augmented Generation (RAG) architecture to generate responses from only the ingested contents.