| Bookshelf Home | Contents | Index | Search | PDF | |
Siebel Analytics Performance Tuning Guide > Siebel Analytics > Add Aggregate Tables >
Basic Guidelines
When you build aggregate tables, do not combine multiple-levels of aggregation. Instead, put each level of summarization in its own table. This simplifies writing queries and maintaining the aggregate tables.
When you create a aggregate table that spans dimensions and contains a large number of rows, create a star schema consisting of the aggregate table and corresponding dimensions. In this way, you can take advantage of the star schema structure and its usual indexing strategies.
For example, suppose you create a aggregate table that summarizes weekly product sales by city. The star schema would include a single aggregate table and three "shrunken" dimension tables:
Each dimension represents a summarization of a detail dimension. For example, the Weekly Dimension table is a shrunken dimension of a more detailed Daily dimension table as Figure 5 shows.
To further improve query performance, you can index the dimension and fact tables depending on the characteristics of their columns.
Too Many Aggregate Tables
Beware of too many aggregate tables. As the workload changes, aggregate tables like indexes fall into disuse. When this occurs, you should remove the aggregate tables. Similarly, should the data distribution shift so that the aggregate table fails to compress data adequately, you should remove the table.
Storage Space
Aggregate tables are effective and they do occupy physical storage space.
| Bookshelf Home | Contents | Index | Search | PDF | |
Siebel Analytics Performance Tuning Guide Published: 18 April 2003 |
