Solving Operational Business Intelligence with InfoSphere Warehouse Advanced Edition

Chapter 5. Temporal data management and analytics in an operational warehouse 179

This type of temporal capability typically includes the following use cases:

򐂰 Auditability - What records of a table have changed in the data warehouse,

and when did they change?

򐂰 Recoverability - Can we restore a table record to a previous state of values at

a particular point of time in the past?

򐂰 History - Related to auditability and recoverability, can we view a prior state of

values of a table record at any time in the past?

򐂰 Traceability and governance - Did a particular change to a value in the

database record occur at the correct time, or has a series of changes across

tables occurred in the correct sequence?

Because operational data warehousing deals with frequent inserts and updates

of the data in the system, this kind of temporal awareness is particularly useful to

manage data consistency, especially for sensitive data.

5.1.2 Temporal concepts

From the preceding discussion about use cases, several repeating concepts

have emerged. These concepts are defined in summary here and are further

developed in the discussion of temporal awareness in InfoSphere Warehouse

Advanced Enterprise Edition 10.

Business time

Business time refers to a date or time stamp that has particular value to a

business process or in a business context. Validity dates and effective dates for

policies, activities, or any business process are example uses of business time.

Business time is more interested in when something happens, will happen, or is

valid in the real or business world, rather than when something happens in the

database system. For example, business time is interested in when a lease

agreement is in force, not when the lease agreement was entered into the

database.

System time

System time refers to a time stamp of precisely when a database record was

inserted, updated, or deleted in the database system itself. System time is more

concerned with the exact moment when any kind of data change occurred in the

database (the database transaction) independent of the business context of that

change. For example, system time is interested in when a transaction occurred

requesting that a customer’s mail be held while they are on vacation rather than

the dates that the mail will be held.