Sponsored content from Broadcom

“Automate everything” initiatives are everywhere. Yet, despite this intense focus by companies across all industries, the actual work of automating tasks, processes, and workflows is not proceeding at anything close to the speed and scope that businesses need. While automation is the undeniable goal, there are indisputable hurdles when automating operations.

The difficulties with automating IT operations are many. There is the skills gap that plagues IT departments. There is the challenge of balancing costs versus levels of service. There are resource constraints since the same people being asked to automate processes are also tasked with “keeping the lights on.” Even after certain activities have been automated, the work does not stop: people have to monitor and maintain existing automation since there are always changes taking place in both business and IT environments.

While these hurdles all exist, they are not insurmountable. You can accelerate the development and deployment of new automation by taking a structured, skills-based, sustainable, and scalable approach to how you automate.

1. Structured IT Automation

To automate IT operations effectively, you need to approach automation in a way that is structured and rigorous, leveraging what you have and working intentionally to make further automation easier in the future. Failing to do so — such as by choosing a siloed approach to automation or by applying automation “band-aids” whenever a department makes a request — will simply add complexity in the long run, further frustrating and delaying your automation objectives.

Creating a structured plan for automating your IT operations encompasses several aspects:

• Automate selectively. Determine which tasks you want to automate to reduce manual effort and increase efficiency. These tasks can include IT operations activities ranging from performance monitoring and reporting to data correlation and analysis.
• Automate strategically. Define specific purposes, permissions, and limitations for each automation to prevent triggering unintended consequences. After you specify your intended outcomes, you can assess how to best achieve those outcomes within these guardrails.
• Automate incrementally. Do not bite off more than you can chew at first. Remember that repetitive tasks might take only five or ten minutes to complete manually, but automating those minutes rapidly adds up to significant time saved for more value-added work.
• Automate intelligently. It is one thing to automate a single task. It is another entirely to automate that task in such a way that the command can be easily and seamlessly re-used in multiple automation sequences. By applying a forward-thinking mindset to each automation activity, you make it possible to accelerate future automation efforts.

2. Skills-based Automation

To accelerate your automation initiatives, it is important to decouple two terms that are often referred to interchangeably: automation and orchestration. Technically, “automation” refers to programming a single action to take place. And, “orchestration” is when you link multiple automated actions together in a sequence to achieve an outcome, such as optimizing a workflow. Frequently, when people say they are automating something, they are actually involved in orchestration.

This may seem like splitting hairs, but the distinction is critical because the work involved in automating tasks requires a different set of skills and knowledge than the work required to orchestrate a process. By decoupling these activities, you can better leverage your resources for maximum productivity and effectiveness. Some team members will be better suited to creating a library of well-constructed, re-usable automated tasks that are readily accessible to anyone to use. Other team members will have the expertise to stitch those tasks together to drive desired business outcomes. By applying a skills-based approach to your automation initiatives, you will save time and cost as well as increase the quality of your operations.

3. Sustainable Automation

Orchestration is reasonably straightforward as long as the same technology is used up and down the stack. But, what if multiple technologies are involved? For example, what if you want to link together an event planner automated on one piece of technology and a scheduler automated on another? Chances are that they do not “talk” to each other well. If you want to avoid re-creating the wheel for each type of technology in your stack, you have to take a sustainable approach to automation — and that means exposing scripts via standards-based, open APIs.

Exposing automation commands as APIs makes them readily re-usable across different tools and technologies. APIs give your programmers a great degree of freedom to architect and implement automation solutions, and combined with a library of automated tasks, they can incorporate automation best practices regardless of what technology tasks were originally programmed on or for. APIs also support agility, since they allow enterprises to change technologies in response to business drivers without destroying the automation and orchestration that has already been put in place.

4. Scalable Automation

For your automation initiative to be ultimately successful, it needs to be scalable. Automation, after all, is a journey, not a destination. There will always be more to automate as business, technology, and environments change. If you find that your resource needs are scaling linearly with the amount of automation you are developing, you are in trouble. It is imperative to be able to keep up with demand ... without constantly adding more people.

The way to ensure scalability is to take a “divide and conquer” strategy to automation. You want to implement layered orchestration efforts where high-level orchestrators determine which sub-processes are required to accomplish a desired outcome, then delegate those sub-processes to lower-level orchestrators to actualize. Approaching automation in this way keeps everybody working at optimum efficiency, avoids duplication of efforts, and prevents unnecessary complexity from creeping into workflows. In doing so, you gain the ability to deliver automation at high velocity when and where it is needed.

Summary

A structured, skills-based, sustainable, and scalable approach to how you automate will enable you to incrementally build trusted automation into the very fabric of your enterprise. I encourage you to explore Automic^® Automation Intelligence and OPS/MVS^® Event Management and Automation, and also check out the third episode of Broadcom’s AIOps Mini-series to learn more about pragmatic approaches to increasing automation within your mainframe operations.

Michael Kiehl is a senior manager of product management at Broadcom. Michael has been working with the mainframe for over 22 years. With significant experience as an automation developer, he currently leads the AIOps and Automation Product Management team for the Mainframe Software Division at Broadcom.