The term row header compression applies only to column-partitioned tables and column-partitioned join indexes, and only when their data is stored in COLUMN format. Row header compression is a lossless method.
Vantage packs column partition values into a container up to a system-determined limit. When that limit is reached, Vantage begins packing column partition values into a new container.
The column partition values packed into a container must be in the same combined partition to be packed into a container. The row header occurs once for a container, using the rowID of the first column partition value as the rowID of the entire container, instead of storing a row header for each column partition value. Vantage can determine the rowid of a column partition value by its position within the container.
If a large number of column partition values can be packed into a container, row header compression can greatly reduce the space needed for a column-partitioned object.
If column partition values are so wide that only a small number can be packed in a container, the space needed for a column-partitioned table or NoPI join index may be larger than the space required for the same object without column partitioning. ROW format may be more appropriate than COLUMN format for storing the object.
If only a small number of column partition values can be stored using ROW format because only a small number of column partition values occur for each combined partition, the space needed for a column-partitioned object may be larger than the space required for the same object without column partitioning. In the worst case, the space required to store a column-partitioned object can increase by up to 24 times.
- Alter the row partitioning to produce more column partition values per combined partition.
- Remove column partitioning from the object definition.