Mainframe Job Activity Tracking via Replicated DASD
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In mainframe computing environments, it is challenging to accurately record and maintain the timing of job and dataset activities, especially during replication, as data processing does not stop at a specific point in time, leading to outdated replicated copies and difficulties in identifying impacted datasets and jobs during disruptions.
Innovation Solution
A system and method that generate and store records of job activities, including initiation, dataset openings, and terminations, in a job activity record storage area, using SMF records and open exit routines to track vital jobs and datasets, ensuring consistent and up-to-date information across replicated storage systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If replication is performed for each individual storage system, then data consistency across storage systems is improved, but the complexity of tracking and managing replication timing increases
Solution Approach 1:
The patent introduces an intermediary recording mechanism that captures job and dataset activity timing information separately from the replication process itself. This intermediary record allows multiple storage systems to be replicated without requiring complex coordination, as each system can independently reference the recorded activity timeline to determine what needs to be replicated.
Solution Approach 2:
The system performs preliminary recording of job and dataset activity timing before replication occurs. By capturing the state of data access and modification activities in advance, the replication process can proceed without needing to stop or coordinate with ongoing data processing, thereby maintaining consistency while reducing management complexity.
2Productivity
If data processing continues during replication, then productivity is maintained, but the accuracy of replicated copies deteriorates as they become outdated
Solution Approach 1:
The patent implements a feedback mechanism where the system continuously records job initiation, dataset opening, and job termination events. These recorded events provide feedback information that allows the replication system to identify which datasets have been modified since the last replication, enabling accurate incremental replication without stopping data processing.
Solution Approach 2:
The system dynamically adjusts replication based on recorded activity. Rather than using a static replication schedule, the system uses the recorded job and dataset activity timing to dynamically determine which data needs to be replicated, allowing replication accuracy to adapt to changing data access patterns while maintaining continuous processing.
3Reliability
If job activity records are maintained for the duration of replication events, then disaster recovery capability is improved, but storage requirements increase
Solution Approach 1:
The patent extracts only the essential timing information (job start/end times, dataset open/close times) from the overall job activity. By recording only these critical temporal markers rather than complete job execution details, the system maintains sufficient information for disaster recovery while minimizing storage requirements.
Data Source
AI summary
A system writes to a replicated direct access storage device (DASD) a record of each step within a job as each step begins and as each step completes. The records are maintained on the replicated DASD for a predetermined period of time. The predetermined period of time is, for example, the greatest amount of lag in replication of all storage systems operating within the system. The records are stored, for example, in an open jobs and datasets (OJD) file, where the file itself is a dataset. The dataset is written to by an online task (e.g., OJDSTC) which gathers input from two sources. Upon job completion, the records are stored, for example, in an OJD journal and removed from the OJD file.