Automation, optimization and modernization of batch system management
Batch processing accounts for most peak mainframe workloads at large enterprises, providing essential back-end digital capabilities for mobile, cloud and web applications. As demands on these back-end mainframe batch processes intensify in terms of scale and performance, the pressure to ensure compliance with SLAs and control costs increases. Further, the responsibility for batch management is rapidly shifting from platform veterans to Millennials with less experience, putting enterprises at risk of critical operations failing, being delayed, or generating excessive costs.
Compuware ThruPut Manager automatically and intelligently optimizes the processing of batch jobs. By balancing workload and improving batch throughput, ThruPut Manager delivers significant savings, including software license fees. Key benefits include:
- Immediate, intuitive insight into batch processing that even inexperienced operators can readily understand
- Better prioritization of batch processing based on business-based policies and goals, automatically selecting the most urgent jobs first without system overload
- Increased data center control and proper batch execution by verifying that jobs have all the resources they need and proactively managing resource contention between jobs
- Dramatically reduced IBM Monthly Licensing Costs (MLC) by minimizing rolling four-hour average (R4HA) processing peaks without counter-productive “soft capping”
- More mainframe resources available for mobile, web and cloud applications since better managed batch workloads reduce overall demand
Automated Service Level Management
Provides an automation framework for batch SLA management, including service goals, reporting and automated actions to ensure SLAs are met. Real time information is supplied on current service levels and trends.
Workload Balancing and Throughput
Dynamic initiators automatically control batch load and distribution to prevent over-initiation and improve batch throughput. Operations are more controlled and efficient. System performance evaluations are taken every 10 seconds, enabling precise decision-making amidst changing business demands.
Workload Prioritization & Escalation
Automatically re-orders the batch queue ensuring that the most important workload gets done first and is progressively escalated as SLAs become more critical.
Automatic Rule Retention
Once set up, automatically retains the business rules for batch workload for the installation. Rules are then applied consistently and repeatedly to system performance, improving batch throughput and maximizing savings. Rather than making decisions under pressure, Ops teams instead make thoughtful choices ahead of time.
Automated MSU Reductions and MLC Savings
Automatically monitors R4HA usage and reduces MSU consumption to lower MLC monthly software charges when sub-capacity pricing is used. Reduced MSU consumption with ThruPut Manager is achieved through constraining and deferring workload, rather than manipulating cap levels.
Workload Demand Constraint
Automatically constrains batch demand as peaks in the R4HA approach the cap to lower MSU consumption across LPAR Groups by an average of 10%.
“Safe Capping/Soft Hammer” Approach
Enables the datacenter to avoid the negative effects of “hitting the ceiling” when R4HA caps are in place. Automatically reduces demand in advance of the cap, constrains demand during the capping period(s) and automatically restores workload to normal operations once the cap has passed.
Datacenter Standards Management
Automatically assigns job parameters to conform to datacenter standards, including job class, priority, job name, account, SYSOUT parameters, WLM service class, and the scheduling environment. Replaces Mellon Mods and many in house JES2 user mods and exits.
Job Resource Analysis
All jobs are analyzed at submission to determine their resource requirements (up to 250 unique identifiers) to facilitate automated batch management and workload control.
Job Routing Management
Automatically determines resource dependencies (e.g., licensed software, etc.) ensuring that jobs run only where and when the specific resources are available. No user JCL changes are required.
Dataset Contention Management
Eliminates dataset conflicts and ensures higher priority workload gets first access by detecting and resolving dataset contention.
Virtual Volume Staging and HSM Recall
Performs essential setup requirements while jobs are waiting in the queue.
Enhanced User Controls
Dependent job control, date/time based job selection, and enhanced SDSF displays provide users with additional job control and status info.
What our customers are saying:
…on z/OS Software License Savings
“ThruPut Manager paid for itself in the first year in software license savings alone.”
“We saved $1,700,000 in the first year with software maintenance cost reduction.”
“ThruPut Manager paid for itself in the first 3 months with software license savings.
“We recently lowered the CPU capacity in one of our busiest LPARs by 10%, thus reducing the MSU footprint and cost. By utilizing ThruPut Manager, we were still able to deliver our batch SLA. In fact, the results have surpassed our expectations. MVS Solutions provided excellent customer service and guidance during the implementation.”
…on z/OS System Resource Constraints
“Initiators are much better utilized; on two systems a 25% reduction was seen, which means getting more work done faster.”
“Helps improve performance of batch processing.”
“More than a 50% reduction of jobs waiting in the queue.”
“ThruPut Manager virtually eliminates initiator juggling.”
“Good balance between not having the system over-initiated and not having a backlog of jobs awaiting execution.”
“Everything runs in the same job class. This saves us time by eliminating the need to manage the JES2 initiator structure.”
“We run more batch with less resources. We are very pleased.”
…on z/OS Batch Automation
“Much less user JCL maintenance.”
“Very significant dollars saved in software licensing costs as well as effective workload balancing.”
“This makes unattended operations possible.”
“We like the idea that ThruPut Manager will modify service classes as a job is running, making this hands off for the operators.”
“My users have never had so much feedback on service performance. It accounts for every reason their job may have been delayed, and they really appreciate the insight.”