From Chaos to Clarity: How One Report Triggered a Workload Automation Golden Age
Every Workload Automation professional knows the feeling: you see the inefficiencies, the recurring system failures, and the wasted processing cycles, but getting business teams and busy management to care about batch environment hygiene can feel like an uphill battle.
Data is only as powerful as the narrative you build around it. This is the story of how a single native tool—Report 01 (Job Run Statistics)—paired with strong naming conventions, transformed my company's operations overnight.
The Baseline: Clear Accountability Meets Visible Failures
In my previous role, we had established airtight, robust naming conventions across our entire scheduling ecosystem. Because our job names explicitly mapped to functional business lines, anyone—from line engineers to executive directors—could look at an active job definition and immediately know exactly who owned it. This level of environmental health meant we had total transparency, but it also meant there was nowhere left to hide when things went wrong.
I was handed a daunting directive: dramatically lower the corporate abend rate. Instead of over-engineering a complex monitoring solution or burying myself in customized scripts, I turned directly to a core diagnostic weapon in our scheduling toolkit: Report 01.
The Strategy: The 90% Abend Filter
If you run a broad performance report across an entire production ecosystem, the sheer volume of data will cause your audience's eyes to glaze over. Management doesn't have time to parse 500 pages of healthy green text. To make an impact, I needed to isolate the critical noise.
I leveraged the selection filters of Report 01 to isolate only the absolute worst offenders. Specifically, I filtered the output to show jobs with a historical failure frequency where:
$$\text = \left(\frac\right) \times 100 \ge 90\%$$
This aggressive filter shrunk a massive, unmanageable data set into a hyper-focused, concise, single-page list containing only a handful of jobs. These were the worst actors in our entire environment—jobs failing nine out of every ten times they were triggered.
The Meeting That Changed Everything
I printed multiple copies of this condensed report and brought them directly into our cross-functional management meeting. I handed them out to everyone at the table.
The reaction was immediate and absolutely brutal. Because our naming conventions clearly broadcasted job ownership, managers looked down at the paper and immediately identified exactly whose teams were slipping. They began questioning each other right there in the open room, demanding to know how anyone could let their applications abend dozens of times without intervention.
Before the next scheduled meeting, the tables turned completely. My desk was flooded with managers who previously paid no attention to batch schedules. They were suddenly lining up to ask what their top-abending jobs were and requesting automated reports to be emailed straight to their inboxes. They were suddenly deeply invested in the very thing I was deeply passionate about.
By highlighting accountability, we systematically eradicated the chronic abend problems. But it didn't stop there. Once those fires were put out, that newfound operational visibility propelled the entire company into what I call the Golden Age of Scheduling. We transitioned from reactive firefighting to building highly efficient, highly complex, and exceptionally stable automated workflows. This single report was the catalyst that changed our entire corporate culture.
Technical Deep Dive: Deconstructing Report 01 (Job Run Statistics)
To replicate this success in your own automation community, you need to know exactly what this report tracks and how to manipulate its parameters to drive accountability.
Report 01 collects historical job execution data over a specified time frame to deliver critical performance metrics. Here is how it functions under the hood:
Scope & Primary Queries
The report allows you to filter by Job Name, Workstation Name, and User Login. Crucially, each of these fields supports standard wildcard syntax, allowing you to easily capture entire application blocks (e.g., FIN_* or HR_PROD_*).
- Important Constraint to Remember: The report does not evaluate or include workloads that were submitted using an alias name. To maintain data integrity, enforce standard scheduler definitions.
Exception and Performance Metrics
Report 01 calculates exact percentages of jobs falling into specific execution states:
- Success vs. Error rates: (The raw total of runs divided into Successful and Error states). Look at a real example from a production dump: a job like /ORCHESTRATE-DEMO-JOB-ABEND01 will stick out like a sore thumb with 16 total runs, 0 successes, and a 100.00% Error rate.
- Total Volume vs. Total Reruns: (The core metric used to find unstable workflows). For instance, a file transfer script (/JNCB-FILE-TRANFER) showing 6 runs, 6 errors, and a rerun flag instantly exposes an infrastructure bottleneck or a bad credential string.
- Runtime Exceptions: Tracks the precise percentages of jobs that Started late, Ended late, or hit Long duration thresholds.
========================================================================
SAMPLE REPORT BLOCK: RUNS BY STATUS
========================================================================
Job: /IWA-MASTER_1#/ORCHESTRATE-DEMO-JOB-ABEND01 [cite: 71]
Script: /opt/WA/scripts/leon1.ksh [cite: 77]
Status Number of Runs % of Total Runs
———— ————– —————
Successful 0 0.00% [cite: 80]
Error 16 100.00% [cite: 80]
Total 16 [cite: 80]
Total Reruns 0 [cite: 80]
========================================================================
Optimization and Granular Auditing
Beyond high-level percentages, you can dive deep into individual job properties:
- Job Details: Extracts the logon user, job creator, hardcoded script location, description metadata, and built-in JSDL XML definition properties.
- Duration Baselines: Identifies the minimum, maximum, and average elapsed duration. (Note: These average and boundary calculations apply to successful runs only, giving you a true, clean optimization baseline).
- Infrastructure Consumption: Provides exact CPU consumption metrics (note: this feature is exclusive to UNIX/Linux workstations and is not available on Windows environments).
Output Customization
The report offers flexible layout presentation options depending on your target audience:
- Charts View: Ideal for executive summaries and high-level dashboarding.
- Table View: Best for engineers who need to drop data into spreadsheets for deep-dive parsing.
- TOC Organization: For multi-department operations, you can automatically format and include a clean Table of Contents grouped either by job or by workstation.
Final Thoughts for the Community
A software tool is only as powerful as the operational standard backing it up. Report 01 provides the raw data, but combined with solid naming conventions and a calculated presentation, it becomes an instrument for organizational change.
If you are struggling to get traction for your workload optimization projects, stop sending massive text logs. Filter down to your worst 90% failures, identify who owns them, and bring that clarity to the table. This report can help you stop fighting fires and start moving forward.
