Glossary

The following list of terms may help you understand the terminology related to CMDB, its activities, processes, and so on.

Atrium Integrator

The Atrium Integrator is an integration engine with which you can transfer data from external data stores to BMC Helix CMDB classes.

CI (configuration item)

This could be any asset or device in a company such as a computer, computer peripheral, server, air conditioning unit, software application, and the organization itself.

dataset

Collection of data that is grouped together by a common value that is set in the DatasetID field of a CI.

discovery applications

Discovery applications help identify various systems in the network and obtain information about them that is relevant to the CMDB. Applications such as BMC BladeLogic Client Automation Configuration Discovery provide data to the Integrator. Applications such as BMC Discovery provide data directly to the Import dataset.

federated data

Federated data is data stored outside BMC Helix CMDB, but linked to CIs so that it is accessible through BMC Helix CMDB. The most common types of federated data are related information and detailed attributes.

identification

Process that identifies records where ReconciliationIdentity (REID) is equal to zero. Any CI where REID is not zero is considered as identified.

identification group

Set of identification rules that qualify data to acquire non-zero REID from the target dataset. target dataset is the dataset into which the record will be merged or inserted during the Merge activity.

import dataset

A dataset is a local grouping of data. An import dataset is data in its unprocessed form and cannot be used for ITIL processes until it is normalized and reconciled.

Managed CIs

CIs that are populated in CMDB as a result of an automated process such as CIs brought in from BMC discovery.

merge

Merging of data from two or more records into one production record. Typically BMC.ASSET.

normalization

Normalization is the imposition of standards or regulations as defined in the product catalog.

The normalization engine performs the following tasks to standardize, correct, and clean up the data drawn from various discovery sources:

Updates the name and version information for products and suites based on the definitions in the product catalog.
Sets impact relationships that define dependencies between CIs.
Sets attributes that are customized for specific use cases.

orphan

A record that would typically not be found in the CMDB and related with a 'hosted' type of relationship, unless it is hosted on a machine and imported as a related item to that host. For example IpEndPoint is a hosted CI. This would also be called a weak related CI.

precedence group

Set of weights assigned to a particular data value that is considered more valuable when compared to other data.

precedence set

Collection of Dataset references where each dataset is associated with a weight assigned to it by the associated precedence group.

production dataset

This reliable data helps you understand the environment and the impact to key business services when changes occur in this environment.

Reconciliation

Reconciliation is the process in which data from different discovery sources is checked and corrected to maintain consistency while also making sure there is no duplication of data.

ReconciliationIdentity

This is an attribute which is assigned to a configuration item (CI). Its purpose is to streamline identification of records as they propogate from data sources into one production dataset.

Regular CI

CIs that are created directly under the regular classes as mentioned in the Common Data Model.

relationship

This is a property that defines how two CIs are related. A relationship is often normalized to have a common name, impact direction and cardinality. BMC's CDM (common data model) predicts the relationships between common class IDs and attempts to restrict types of relationships between classes. Relationships can also be 'weak' although no one refers to a relationships as 'strong'. Relationships are either weak or not weak.

sandbox dataset

Any changes that need to be made to the production dataset are first tried out in the sandbox dataset. If the implementation is successful, only then are they moved to the production dataset. The sandbox provides a safety mechanism that prevents unintended changes to your production dataset.

soft deleted

CI or relationship that is marked for deletion in the next purge activity. Any record that has the value of MarkedAsDeleted as Yes (MAD.Y) is considrered as 'soft deleted' and any CI that has been deleted from the database is known as hard deleted.

UDDI (Universal Description, Discovery and Integration)

This is a platform independent framework functioning like a directory that provides a mechanism to locate and register web services on the internet.

Parallel hierarchy

A copy of the existing CDM hierarchy. The denormalization changes are first implemented in the parallel hierarchy.

Refresh class

Classes that are immediate children of the denormalized classes.

New class

Classes in the parallel hierarchy that are denormalized by converting to categorization sub class.