Ahead with innovation!
I think workload automation is probably one of the biggest challenges we must engage if we want to move with the times in information technology. I strongly believe we are in front of a major innovation in our IBM Workload Automation solutions developed by our partner HCL.
When I presented this topic for the first time, and when I finished talking one of my colleagues, @OzgunOdabasi270002TRU1, said something that I really liked. He said, “Why don’t you begin from the end with your presentation?” I started thinking about this and eventually convinced myself that Ozgun’s suggestion would be totally effective! Therefore I decided to start this series of blog posts from the conclusion, so let’s begin by looking at this picture:
I am pretty sure some of you saw this and thought, “Oh my goodness!” This thought was probably the same whether you work in the mainframe or distributed systems area. The first thing we notice is the name of the file, JOB3FTP, which makes us think about a file transfer job even without specific skills. The second thing is the content, which requires some skill to understand. This is a file transfer batch job, startable by an IBM z/OS system, written in XML and designed to be executed outside of z/OS.
It isn’t easy to find people with concurrent mainframe and XML skill, which generates a question: Who is going to write such a job to be initiated from z/OS and executed outside? A young guy from the university will probably enjoy this kind of work in a z/OS job library, using hexadecimal characters and being careful with continuations, because he is learning something and doesn’t mind this challenge. But the reality in IT is quite different, since the number of jobs to write can be very huge and can be tedious work. So we need a way to automate this process.
End–to-end batch processing
In my next post I will get back to the role of XML, which illustrates the power of the workload automation plug-in innovation. But first it is important first to give a picture of the end-to-end batch processing concept.
Traditionally batch in computer science identifies the ability to run complex processing “in silence,” without any human interaction. This year in IBM we are celebrating the 50th anniversary of the mainframe, and we could also say that batch processing was born in 1964 with the mainframe.
Batch processing has solved a lot of problems and saved people time with their mainframe customer business due to its unattended characteristic (meaning that humans do not have to oversee it). IBM Workload Scheduler (IWS) automates the scheduling, taking into account all the prerequisites, so that there is no need for manual scheduling of batch jobs.
For more than 30 years batch processing was confined to a mainframe, but in the 1990s there was a fast explosion of distributed platforms, from UNIX to Linux, from Windows to proprietary applications, extending the need of batch processing to distributed environments as well.
IBM and HCL offer IWS to completely automate batch processing independently on the target platforms. It automates, monitors and controls workflow throughout the enterprise IT infrastructure.
IWS supports many types of configurations, so a job can run in any type of platform: z/OS, Linux, Windows and so on. So end-to-end batch processing is the ability to start jobs from any chosen platform and execute them on other platforms. The most common end-to-end batch configuration is based on an engine located in IBM z/OS and several agents installed in the target z/OS and distributed platforms. This is IBM Workload Scheduler for z/OS.
Between 1990 and 2000 the IBM Workload Scheduler was for the first time enhanced in a multiplatform direction. That enhancement was a big innovation for that period, because it made possible starting simple scripts or batch commands from z/OS and running them on UNIX, Linux, Windows, OS/400 systems and of course on Cloud.
In the second part of this series, we will continue our discussion on the evolution of IBM Workload Scheduler and further innovations in workload automation. We will talk about the application batch processing workloads, XML and the innovation of plug-ins.
Please stay tuned, and share your thoughts with me on Twitter @nicochillemi.