Understanding Jobstream Analysis

To use AAI most effectively, you need to understand how AAI can change the way you can monitor your batch schedule. AAI offers a unique model for monitoring batch in that it is a predictive tool unlike anything else currently on the market. As such, it allows you to proactively alert your support staff regarding your critical batch processes. In addition, it can be used to distribute monitoring in such a way as to allow selected users a powerful view on their batch that can complement existing centralized support mechanisms.

To accomplish this, AAI introduces several new objects to the scheduler environment. These include:

• Target Job: A Target Job is an Endpoint in your batch that you want to monitor and/or perform historical analysis on. It is literally a scheduler job or box(group) that you use as a criteria to determine lateness. If the Target Job is late, then this in turn becomes the criteria for alerting. Moreover, the Target Job is used by AAI to automatically populate its upstream dependencies in order to generate a jobstream, defined below.

• Jobstream: A jobstream is the Target Job plus any upstream dependencies that may fall onto the critical path. This may include a few or perhaps thousands of jobs and/or boxes (groups). Upstream dependencies that can never fall on the critical path are not considered part of the jobstream.

  Note:

  Jobstreams can only have one Target Job, in other words, only one job or box (group) can be selected to create a jobstream.

• Critical Path: In AAI, the critical path consists of all jobs that have directly affected or are predicted to affect the completion time of a jobstream. In other words, if any job on the critical path is delayed, the jobstream is also delayed. If a job is not on the critical path, delay associated with the job does not necessarily affect the completion time of the jobstream. However, if a job not on the critical path is delayed enough to effect the completion of the jobstream, it will dynamically be added to the critical path in real-time. Therefore, the critical path can change from run to run.

• Scheduler: A scheduler in AAI essentially maps one-to-one with a scheduler instance. A scheduler is defined after AAI is installed by pointing to a scheduler instance and populating the AAI database with the scheduler-specific data. This data, when combined with AAI-derived data that includes delay information, prediction statistics and so on, comprise the scheduler.

These new objects, when combined with powerful prediction engine and alerting mechanism, give you a new approach to monitoring your batch that can save your organization time and money in innumerable ways. After the AAI Administrator defines Target Jobs and jobstreams to AAI , you can use the AAI Analysis Tab to perform the following functions:

• Troubleshoot application bottlenecks that can be holding up jobstream execution

• Determine different types of job latency or jobstream delays

• Check on historical run data for a particular jobstream

• Predict outcomes of currently running critical jobs based on the current and historical metrics

For information about how to add jobstreams and define the related features see the following sections:

• Adding/Editing/Deleting Jobstreams
• Configuring Alerts in Jobstreams
• SNMP Management Information Base (MIB) Details