Impact analysis overview

An impact analysis is based on results identified by the impact analysis engine. The impact analysis engine traverses the configuration management database (CMDB) and locates the CIs that are expected to have outages as a result of target CI outages. Most commonly, it does so by using user-configured rules that are based on CI relationships. During the analysis, you can also instruct the engine to retrieve historically identified impact data.

A Change Analyst conducts the impact analysis, which consists of two steps:
  1. The impacts of a Change on CIs across your data center are identified automatically by the impact analysis engine. This automated impact analysis can be kicked off by the workflow to occur after you create implementation tasks and attach target CIs to the tasks. You can also manually invoke the impact analysis engine using application actions. You might manually run an impact analysis if you did not apply a workflow to a Change; if the automated impact assessment action was removed from the workflow; or if you want to assess the impact of a target CI that you add, an outage value that you Change, and so on.
  2. You examine the results of the automated impact analysis to determine the disruption that might result from a Change. You can perform a number of operations during the analysis: identifying impacts task by task, identifying the impacts of a single target, adding impacts to a task based on historical knowledge, and so on. Based on this analysis, you can recommend for or against approval of a Change, make adjustments to minimize disruption, or otherwise modify the Change.

During the planning stage, an administrator configures the impact analysis rules that are used by the engine. The administrator also configures system properties that determine how the engine calculates historical impact data. The rules and criteria that are set by your administrator are applied when you conduct an impact analysis.

Defining outages

When an impact analysis is executed, the impact analysis engine uses the target CI on the implementation tasks as the starting point to determine what outages will occur when the change is implemented. First, the engine determines whether the target CI will be offline when the change is implemented. If the target CI will not be offline, then no further impact analysis is needed for that target CI. If the target CI is offline for the change or degraded, the engine applies the impact analysis rules to assess the impact of the change on other systems that are related to the target.

When you specify a CI that is the target for a change, you can indicate the type of outage that is expected for that CI when the change is implemented. The following values are available for the Outage field:
  • Degradation - The CI will be degraded.
  • None - Implementing the change has no effect on this CI.
  • Offline - The CI will be offline when the change is implemented.
If no value is specified in the Outage field, the impact analysis engine assumes that the CI will be offline.

Impact analysis results

When the impact analysis engine runs, it identifies all CIs that are impacted. However, because many of the CIs that are identified are not relevant, by default IBM Control Desk reports results only for key CIs. Viewing only key CI results makes it easier to determine which CI outages affect the business and helps you make better decisions about scheduling and risks.

A CI is considered a high value or key CI if it belongs to a classification that is included in impact analysis results. When creating CI classifications, select the Show in Impact Analysis Results checkbox to specify that the classification is included in impact analysis results. If this field is selected, any CIs with this classification are included in the impact analysis results. Note that this attribute is available only for classifications that are used with CIs; that is, the classification specifies CI in the Use With field. As a best practice, restrict the number of classifications that have this value to classifications with high business value, for example, business applications or business services. Some classifications that are provided with the product include this attribute:
  • CI.BUSINESSPROCESS
  • CI.BUSINESSSERVICE
  • CI.BUSINESSAPPLICATION
  • CI.BUSINESSSYSTEM

By default, IBM Control Desk reports results only for CIs with a classification that is flagged for impact analysis results. However, if administrators want all results reported, the filtering feature can be disabled with the system property pmchgia.filter.impacted.cis.

Restricting the types of CIs that are reported in the results does not affect the types of CIs that the engine analyzes. All CIs are traversed and all impacts are identified. This feature affects only the list of impacted CIs that is reported and stored. If an impacted CI meets any of the following criteria, then that CI is included in the impact analysis results:
  • The classification of the CI is flagged to Show in Impact Analysis Results. CIs of this classification have high business value.
  • A Business Impact value is specified for the CI. CIs with this value are used in Risk calculations
  • A Change Window Calendar is specified for the CI. CIs with this value are used in Schedule Conflict Analysis & Scheduler.
  • The CI is listed in the Scope of an active Blackout Period whose End Time has not passed. These are used in Schedule Conflict Analysis & Scheduler.
Parent topic: Conducting an impact analysis