Job automation under z/OS in centralized and distributed IT landscapes is what makes the Discovery Agent Network so attractive. And there are numerous applicaton scenarios out there to prove this.
To make the whole solution even faster and more scalable, our developers just added an auxiliary function to Beta 92 EJM: distributed ‘high-frequency’ jobs with job intervals of under one minute.
Notwithstanding countless prophecies of doom, the mainframe continues to be the key IT tool in large corporations – even though many applications are no longer executed on the mainframe, instead having migrated to Unix/Linux and Windows server platforms. In other words, a shift in task distribution is taking place.
Powerful Helper of the Mainframe Scheduler
Given this trend, companies are facing the challenge of setting up consistent job scheduling across the corporation. Many large companies solve this by combining the IBM Workload Scheduler (IWS), formerly known as TWS, with the Discovery Agent Network provided in Beta 92 EJM, which acts as an innovative extension.
This setup makes it possible to dispatch workloads to distributed servers while still controlling scheduling for SAP and all other server-related processes from the mainframe. The main benefit is that this makes it very simple to connect applications operated under Unix, Linux and Windows with the centralized automation process.
Our customers confirm the efficacy of our distributed agent concept: They can’t find any competing software able to extend the mainframe scheduler as effectively and at the same level of performance as the Beta 92 EJM agent system.
Automating Applications in a z/OS Environment with the Discovery Agent Network (Beta 92 EJM)
Central Scheduler Is Not Unduly Burdened
We recently further brushed up our workload automation portfolio by adding a special function to Beta 92 EJM: so-called high-frequency jobs. This feature speeds up the launch time of distributed server jobs that recur in a given time interval without burdening the central scheduler. The job start itself is triggered by the IWS, but all further actions are passed on to SAP via Beta 92 EJM.
Provisioning of spare parts in the parts warehouse of a car manufacturer serves as a perfect example case. In this kind of a warehouse, there is an incessant in- and outflow of parts. Also, requests for specific parts enter the warehouse on a constant basis, and somebody then needs to find the corresponding part.
Tasks arise in chronological order. In order to process them, the local IT department must schedule jobs in the respective specialist application in a conventional manner and then monitor job processing. The complex task of monitoring the warehouse, order entry and shipping binds a lot of manpower.
Many Slices Add Up to Smooth Processing
The high-frequency jobs delivered by Beta EJM 92 simultaneously launch the various process steps as ‘slices’, with each slice representing a single HF job. The ‘goods receipt’ slice checks the relevant application to determine whether a new part has arrived and books it upon receipt.
The next slice accompanies the pallet carts on their journey through the warehouse, while other slices check the storage process, new orders, tour planning, delivery to the truck, etc. Instead of waiting for their turn in chronological order, many slices are active in parallel – one for each process step. These high-frequency jobs can be executed in intervals of less than 60 seconds.
Savings in Terms of Mainframe Workload and Licensing Costs
Such a high-pace job control scheme will normally place a high load on the IWS if it had to control all jobs. Yet since the Discovery Agent Network runs the jobs centrally, the IWS is not subjected to more load than necessary. This helps to save mainframe resources and, at the same time, facilitates cost-efficient job scheduling.
The new Beta 92 EJM function thus allows operators to automate processes that need to be triggerd and monitored manually when drawing on a conventional solution. The warehouse IT departments benefit from the service by being able to make more productive use of their time. Thinking this one step further, the company’s customers ultimately benefit from the centrally controlled order process, as they get the requested parts much faster. This illustrates how group-wide host job scheduling can have a significant positive impact on the bottom line.