Skip to end of metadata
Go to start of metadata
The Resource Monitor is based on the concepts of monitors and rules.
From an administrative point of view:
- Monitors are automatically evaluated by the Near-Real-Time Warehousing engine while it stores data into the BMC TrueSight Capacity Optimization Data Warehouse.
- Rules are evaluated by a special type of System task, the Resource Monitor task.
Resource Monitor rules can contain the specification of a time filter that must be used to evaluate the alert, so it is important to know and decide when exactly the alert should be evaluated: this is the main purpose of the Resource monitor task.
As soon as a Resource Monitor rule is created, it is associated to a specific schedule. The list of schedules displayed to the user is exactly the list of scheduled Resource Monitor tasks.
Usually, a single Resource Monitor task that runs every day is configured by default. If you wish to allow users to evaluate Resource Monitor rules using different schedules, simply create a new Resource Monitor task and set the scheduling configuration accordingly.
Working of the Resource Monitor
The Resource Monitor works in the following way:
- Data is imported by ETL tasks in a temporary stage table at the data source time-aggregation level.
- The warehousing engine reads the monitor's specifications, that is monitored KPIs and thresholds.
- The warehousing engine then reads the new data from the stage table, labels the data samples of the monitored KPIs as good, warn or poor according to the specified thresholds.
- If the time detail level of the original data is less than five minutes, the warehousing engine aggregates the data in five minutes chunks, adding up label counters. This way, BMC TrueSight Capacity Optimization ensures that no data in the warehouse has a time resolution higher than five minutes, and that all the monitored data samples have been evaluated against the monitor thresholds.
- The warehousing engine also aggregates data at coarser time resolutions (for example, hour, day) and, in doing so, keeps track of how many good, warn or poor data values are present in the aggregation.
- New and aggregated data are removed from the temporary stage table and stored in the proper tables.
- Each Resource Monitor task retrieves the list of alerts it has to manage, along with their definitions and triggering conditions.
- Each Resource Monitor task then compares the label counters and data-related statistics of the monitored KPIs against its alert rules and, if any conditions are met, performs the alert's predefined actions.
The data labeling process is not retroactive: when a new monitor is defined, its labels are only applied to new data.
For more information about the tasks that you can perform, see: