Test Model Characteristic Description or Value
Total number of categories 492,152
Number of non-measure dimensions 5
Number of measures 5: 1 regular, 2 calculated, and 2 after-rollup
calculated
Source data format ASCII (tilde-character delimited)
Number of source files 9: 6 structural and 3 transactional;
multiprocessing enabled for 4 largest
Number of transaction input records 50 million
Size of all source files 2.28 GB
Enabled options crosstab-caching and auto-partitioning (5
passes with 500,000-record size limit)
The following grid shows the model design, as seen on the Cognos Transformer
(Windows) interface:
Transportation
(7) (2) Date (197) (2)
HR Code
(491,376) (2)
Country or
Region of
Origin (286) (2)
Country or
Region of
Export (286) (2)
Transportation (5
categories)
Year (15
categories)
HR Code 1 (1471
categories)
Region (10
categories)
Region (10
categories)
Month (180
categories)
HR Code 2
(19036
categories)
Country (274
categories)
Country (274
categories)
HR Code 3
(75916
categories)
HR Code 4
(123856
categories)
HR Code 5
(271095
categories)
Multiprocessing with Cognos Transformer Server
By upgrading to a multi-CPU server and running multiple instances of Cognos
Transformer in parallel, you can significantly decrease the build time for your
largest cubes. In a model that generates more than one cube, the overall creation
time is reduced to the time taken for the slowest cube to build.
Use th following strategies:
v Give each Cognos Transformer process its own CPU. Because each instance uses
system resources independently, ensure that each one has sufficient memory, I/O
bandwidth, and disk space.
v Provide each Cognos Transformer instance with its own configuration files.
v Do not share the locations of DataWorkDirectory and ModelWorkDirectory among
multiple instances.
v Add an ampersand (&) to the end of the UNIX or Linux command line to begin
your first process in background mode. For example, use
Chapter 9. Guidelines for Optimizing Production Environments 203
To Next Page
Loading...
Need help?
General