|
Index - Major Sections
Home
**InHCc HMIS**
Site Map
Health Economic and Reform
Benefits
Discussion
Data and Data
Analysis
Health Management
Product and Services
References
Team
_______________
Index - Same Level Subject
Index - Child Subjects
| | Introduction
Informational databases
Characteristics
The Informational Databases has these characteristics:
-
Data
Access tends to be ad hoc
-
May
be non-normalization in design
-
May
contain redundant data and be aggregated.
-
Timeliness
is not generally an issue.
-
Access
large volumes of data records
-
Current
data is not as important as in operational databases.
-
Updated
generally as batches
-
Have
fewer more sophisticated users
Design of the Data Warehouse
Normalization
Renormalization
Denormalization means that you are purposely designing
your database so that it is not normalized. This is usually done to maximize
performance or to simplify end-user reporting. In a normalized database a
query may require many joins to process the request. These additional table
joins can be expensive in terms of computer time. More will be said about
denormalization in the topic on Data warehousing.
Renormalization
Techniques
-
Duplicate
data: Duplicate data can reduce the number of joins required to process a
query.
-
Summary
data: Summary data where used often in queries.
-
Horizontal
partitioning: split the table into two or more separate tables at the
record level, such as by category, thus reducing the number of rows per
table accessed. …example – by department
-
Vertical
partitioning: Splitting the table into two separate tables at the column
level, thus reducing the number of columns per table….example –
salary.
Design
Granularity:
Granularity is the
major design issue in the data warehouse environment. It profoundly affects
the volume of data that resides in the data warehouse and at the same time
affects the type of query that can be answered. The volume of data in a
warehouse is traded off against the level of detail of a query. (I-45)
Partitioning:
Refers to the
breakup of data into separate physical units the can be handled independently
(I-55). Partitioning gives the
user finer detail (I-58)
Date
Type of symptoms
Type of results
By geography
By age
etc
|