IBM Cognos Dynamic Cubes

Chapter 1. Overview of Cognos Dynamic Cubes 7

management of Java heap sizes for the Dynamic Query service that is used for Cognos

Dynamic Cubes.

Other considerations in large deployments require careful infrastructure planning. To share a

Cognos deployment across many demanding applications, routing rules across multiple

report servers are required to optimize each query service so that all users have a positive

experience.

1.3 When to use Cognos Dynamic Cubes

Cognos Dynamic Cubes gives customers an additional tool in the query stack to handle a

growing business problem: high-performance analytics over large data warehouses.

Certain conditions make Cognos Dynamic Cubes the correct choice, acknowledging that

other variables, which are not listed here, might influence your choice:

򐂰 Data warehouse with star or snowflake schema

A properly designed data warehouse is a recognized best practice to provide a well

performing experience for users. As such, a properly designed data warehouse is a

requirement to use Cognos Dynamic Cubes.

򐂰 A server with adequate memory

Cognos Dynamic Cubes relies on memory to maximize its potential. Therefore, a server

with enough memory to support the application can help to realize the full potential of

Cognos Dynamic Cubes. A smaller server can be quite sufficient for small applications

such as proof-of-concept (POC), but when you want to query a data warehouse, a proper

server is required. For more information about hardware requirements, see Understanding

Hardware Requirements for Dynamic Cubes from the business analytics proven practices

website:

http://www.ibm.com/developerworks/analytics/practices.html

򐂰 A database with approximately 25 million or more fact table rows

For small databases with fewer than 25 million rows of fact table data, there are easier

ways to accomplish the task of presenting relational data dimensionally. For example, an

OLAP Over Relational (OOR) model provides the ability to explore data dimensionally, and

uses an existing Framework Manager model. However, OOR models cannot effectively

handle data volumes beyond this size, while providing speed-of-thought interactive

analysis performance.

Cognos Dynamic Cubes is optimized to effectively scale up to terabytes of data, using

aggregate-awareness and in-memory technology to accelerate performance.

1.4 System requirements

Cognos Dynamic Cubes includes IBM Cognos Business Intelligence entitlements, as of

release 10.2. The standard system requirements that are described in the 10.2

documentation and conformance web page apply. However, similar to any large application

deployment, more CPU power and memory are required to serve more than a small volume

of data to a small number of users.

Because Cognos Dynamic Cubes relies heavily on in-memory technology to accelerate

performance, consider using a powerful server with a significant amount of memory to realize

the full potential of Cognos Dynamic Cubes.