Skip to main content

A Journey Through Automation on IBM Z

Joseph Gulla kicks off a new series on automation on IBM Z from system to application with a discussion of IBM Z Service Automation Suite.

"IT Trendz" in white against a purple banner, white chat bubble in righthand corner, with connected cubes against a blue background below.

Some time has passed since I took a close look at automation at the system level on IBM Z. In full disclosure, I once made a living automating mainframe operations. After doing some research, I was excited to find IBM Z Service Automation Suite. Its tagline is “automate and control a large range of system elements spanning both the hardware and software resources of enterprises in a sysplex.” This is a good starting point for a new series on automation in IBM Z from system to application. Let me get started. 
IBM Z Service Automation Suite 
The suite is a collection of products that automate and control a large range of system elements spanning both the hardware and software resources in a sysplex. Many of the products in the suite have been growing and changing over the years and are mature yet evolving, for example, Linux on IBM Z is supported.  
Taken as a collection, you can use the suite to do anything you need to do to monitor and control, in a highly automated manner, everything in IBM Z, including hardware, software products and applications. I used the phrase “everything in IBM Z” because I know about the flexibility of the software and the resourcefulness of the system programmers supporting z/OS and all of its related software. They find ways to get it automated often with a grin on their faces.
System Automation
IBM System Automation for z/OS provides a policy-based, self-healing, high-availability solution to optimize efficiency and availability of critical systems and applications. Policy is built from samples and maintained as part of system definitions and tells the software everything it needs to know about the configuration and its software. 
The automation software is designed to reduce administrative and operational tasks, customization and programming efforts. This savings includes automation implementation time and costs associated with Parallel Sysplex automation. Prior to the widespread use of automation software like IBM System Automation for z/OS, some computer operators would monitor system consoles that were full of routine messages looking for exceptions. When IT projects implemented automation, consoles were made quiet through the suppression of routine messages. Only exceptions appeared so that they could be handled. Eventually, the handling of exceptions were automated and operators were the creators of the policy to do so—some had new jobs as automation specialists.
Network and Systems Management
Tivoli NetView for z/OS provides automation and network and systems management. When I taught classes on NetView, I used to remark that it was a tiger with two heads. One head was its automation platform and the other was its network management toolset. These are two different domains, but when you combine them, you get some really useful capabilities like automating network management. This is done, in simplest terms, by trapping messages that indicate a failure and running an automation routine to recover the resource or notify a human. 
Tivoli NetView has a world of capabilities so rich that you could easily spend a decade mastering the many subtle capabilities of the product, including its rich interfaces with z/OS. When you examine how seamlessly NetView works with z/OS, you understand why its automation capabilities are so rich. The scope is from message trapping and responding (every message that runs through the message processing facility) to AT, EVERY and AFTER commands. These commands are used to schedule queries for the problems you miss by relying solely on message-based automation. That automation strategy is another two-headed animal—one that looks for messages with one head while periodically asking “are you OK?” with the other. This duality is a basic construct of automation thinking.
Workload Scheduling
IBM Z Workload Scheduler provides the infrastructure to help automate the execution of batch and near real-time workloads and activities that are running in support of the organization’s business services. This is a big deal and has been one for a very long time. Organizations that run z/OS often need a plan to run hundreds of thousands of jobs, and then need a highly scalable, highly available workload automation solution to get it done. 
This monumental task is best done centrally with a robust tool. If the tool fails or makes it easy for humans to make mistakes, awful things can happen, such as missed payroll or compliance reports or any number of key work products and reports that come from workloads running on Z. IBM Z Workload has excellent materials in Tivoli documentation central, including wikis, demos and tutorials. 
Management Dashboards
IBM Service Management Unite provides a customizable management dashboard, bringing mainframe management information and tasks from different sources into a single management environment. The software is a free-of-charge component available with IBM Service Management Suite for z/OS  and IBM System Automation for z/OS V4 or higher.
Why do I need it or how would I use it? Three main ideas explain this.
  1. It provides a consolidate view of reports so you can recognize and resolve problems quickly
  2. It provides a single point of control where you can take actions providing integration between different products that have a common purpose. From it, you can issue commands so it’s not just about monitoring, but management. 
  3. It’s prebuilt but customizable so you get up and running quickly and can change it to meet your needs
What’s Next?
I hope this brief trip through automation on IBM Z was a good starter.  Next week, I’ll write about automation of the layer above the OS, such as CICS, IMS, Db2 and others.  

Delivering the latest technical information to your inbox.