Skip Headers
Previous
Previous
 
Next
Next

Cube Storage Formats

You can store a cube in either a compressed or an uncompressed format. The default format is compressed, which is appropriate for most real-world data, which is very sparse.

What Are Compressed Cubes?

Compressed cubes are used to store sparse data. Compressed storage uses less space and results in faster aggregation than uncompressed storage for sparse cubes.

Some aggregation operators are not compressible because their values can change for every level: all weighted operators, all scaled operators, and the hierarchical weighted average operator. When these operators are used in a cube, the aggregation engine compresses the dimensions that it can, but it cannot compress the entire cube.

All measures in a compressed cube must use the cube aggregation rules.

What Are Uncompressed Cubes?

Uncompressed cubes are used to store moderately sparse and dense cubes. They have more options in their configuration than compressed cubes, and these choices can dramatically affect performance.

The measures in the cube can use their own aggregation rules; they do not have to use the cube rules.

The dimension order is extremely important in an uncompressed cube, because it determines the way the data is stored on disk. Performance is optimized when values that are accessed together are stored together, because fewer pages must be swapped in and out of memory. These are the basic guidelines:

OLAP automatically determines the appropriate storage for compressed cubes.

What Are Global Composites?

A composite is the physical structure used to store the data in an analytic workspace. An unpartitioned cube always has one composite for the cube, whether it is compressed or uncompressed. A partitioned compressed cube always has a composite for each partition. A choice between single (global) and multiple composites is available only for uncompressed partitioned cubes.

By using a global composite, you may be able to:

A global composite may be a better choice under these circumstances:

When in doubt, do not create a global composite. The cube will have one composite for each partition.


Note:

A global composite does not allow parallel aggregation across partitions. If the analytic workspace has fewer cubes than available processes, then aggregating the partitions in parallel may improve build performance more than a global composite.