A metric defines the service level objective for a specific service during a specified time period. Contract performance is measured by applying the metrics that have been defined for the contract, and analyzing the aggregated data for a predefined tracking period to determine whether the service level provided matches the level agreed upon in the contract. Metrics made up of several parameters can be customized for each contract according to the contract objectives by modifying the objective statement, parameters and other customizable features.
Relationships between metrics can be defined to enable drilling down from the reports of one metric to the reports of related metrics.
Metrics apply to specific domain categories, for example, Minimal Uptime, Average Bandwidth or Help Desk Response Time. The domain category determines the semantics of each metric, including the unit of measurement and the meaning of deviation from the service level target. Domain categories for related domains are grouped into service domains.
The semantics of a particular domain category are always the same regardless of the contract party or metric, but the way you measure the service level and what you measure may be different. The commonality is the meaning of the result of aggregated value versus the service level objective.
Metrics can also be used for purposes other than representing service level objectives. For example, they can be used for calculating operational data or as an intermediate calculation for penalty formulas. In addition, metrics can be used to test business logic modules.
For more information on how to use metrics to test business logic modules, see Using Business Logic.
|
Copyright © 2012 CA.
All rights reserved.
|
|