|
|
|
Monitoring attributes are defined to every IP resource. Monitoring attributes let you test the performance of a resource against a value or a calculated baseline to trigger alerts and actions.
Note: If you specify a type of baseline alerting, it is based on the hourly rate for total and counters.
An attribute can be one of the following types:
Displays a numeric value that may increase or decrease in an allowable range. For example, processor memory usage varies between zero and the physical limit of the hardware.
Displays the rate of increase in units per hour in a sample period. The rate is derived from sample data, which is an accumulated count that increases in value over time (for example, bytes received). This type is also referred to as COUNT.
Displays the value from a defined set of discrete values. For example, the state of a device can be ACTIVE or INACTIVE. Multiple values are aggregated so that a percentage of a particular value over time is available. This type is also referred to as ENUM.
Displays the rate of increase in units per hour in a sample period. The rate is derived from sample data, which is the total increase in value in a sample period.
Baselines are sliding averages of hourly values. You can use them to monitor network activity more easily.
If you use sampled data, the region calculates baselines for numeric type attributes—gauges, counters, and totals. You can use these baselines for alerting. When you specify a resource's monitoring, you can update the alert details and specify the type of baseline to alert on.
When individual samples are taken, they are compared to the baseline value as follows:
The equivalent hourly rate is calculated and compared to the baseline value. If the equivalent hourly rate differs from the baseline value by more than the specified percentage, an alert is raised.
The original sample value is compared to the baseline value. If the sample value differs from the baseline value by more than the specified percentage, an alert is raised.
When samples are summarized for each hour, the region recalculates the baseline as a moving average and stores the updated baseline value.
All baselines are averages of hourly summary values. The hourly summary values that are averaged depend on the baseline type.
The following baseline types are available:
Averages the hourly summary value for one specific hour of one specific day name, for up to the last ten weeks, for example, Friday at 17:00.
This type is the most granular baseline because you are comparing only the same specific time of the week.
One hundred and sixty eight hour-of-day baselines exist, one for each hour of each day name (24 hours a day, 7 days a week).
Averages the hourly summary value for all hours in a specific day name (that is, Monday, Tuesday, and so on) for up to the last ten weeks.
Day of week is not as precise as the hour of day baseline. Every hour on Tuesday is averaged together. No account is taken for the fact that some hours, for example, working hours, frequently have different workloads than others (for example, late shift or off peak hours).
Seven day-of-week baselines exist, one for each day.
Averages the hourly summary value for all hours in the day, for each day up to the last 30 days.
Daily is the least precise baseline. This baseline does not account for the workload characteristics of different hours. Also, it disregards the different daily workload patterns.
One daily baseline exists.
Note: Baseline values become available after the first hourly summary of an attribute; however, when you start performance monitoring for the first time, baselines are calculated on few hourly summaries. To avoid spurious alerting, we recommend that you collect baseline date for a few days or weeks before you use it for alerting.
This option lets you delete baseline data for an individual attribute and qualifier, thus resetting the baseline values to null. You may want to do this after atypical network activity or when changing a region from testing to production.
Important! Deleting the baseline data affects the current monitoring for the attribute and may stop alerts from being automatically reset. We recommend that you manually close any open alerts for the selected attribute and qualifier.
To delete baseline data
The Security and System Services : Primary Menu appears.
The Data Framework : Baselined Resources panel appears.
A confirmation message appears.
The baseline data for the selected resource is deleted.
| Copyright © 2012 CA. All rights reserved. |
|