Cognos 10 comes with many new features and enhancements that can be leveraged to create cubes that meet your business needs. These include Dynamic Cubes, PowerCubes, and TM1 control cubes. But before you jump right into creating cubes with Cognos 10, let’s take a look at some of the key concepts that help you understand the basics.
Sparsity
In a cube, there are cells or “x-y coordinates” that represent information or data. These coordinates are a way to identify how you want to organize that data, so that you can use it to report on it.
This is what enables you to do things like add maximum, minimum, average, count, and total values to that data. That’s the reason why a lot of people prefer to work with cubes instead of spreadsheets or databases for business reporting and data analysis purposes.
It also helps to know what kind of information you’re working with, so you can determine how to best structure your cubes. If you’re capturing large amounts of data, for example, it’s important to design the cube to allow you to access it as much as possible without slowing down the system.
Choosing the right data source
If you’re going to be using Cognos 10 as your reporting and analysis platform, you need to choose an appropriate data source for the kind of business information that you have. The choice will have a major impact on the scalability and performance of the cube.
For instance, if your business requires a cube that has millions of records, you’ll need a highly scalable data source. Ideally, you should have a data source that can store billions of rows, as that’s the amount of data that is typically reported on and summarized by businesses today.
Choosing an appropriate data source isn’t always easy, especially if you have multiple types of data and need to ensure that all your users can access the information they need. You need to consider the database size and the data storage requirements as well as how much processing power is required by your business users.
You’ll need to decide whether you need to load your data from a structured relational database, or if you prefer to load it directly into an in-memory OLAP (online analytical processing) cube. You can use both technologies, and your decision will depend on what is best for your specific business needs.
Another consideration is the caching of your data and calculations. Cognos has a sophisticated data cache that improves the performance of reports, as it stores aggregate calculations and intersections in memory. This caching is also available in Dynamic Cubes.
The caching is controlled by a query service admin task that can be scheduled to run on a trigger or at a predetermined time. It has a default cache refresh of 24 hours, but this can be increased to accommodate high data volumes.
This caching is also very useful for enabling a dynamic cube to run on multiple servers, providing a scalable architecture that allows the same cube to be deployed across multiple machines. This means that if one server goes down, the other will still be running and able to handle queries.
