Storing data – column storage

Having learned about the row store engine of SAP HANA, now let us learn about the column store engine. Data will be stored in RAM, similar to the row store engine. The concept of column storage has emerged from Text Retrieval and Extraction (TREX). This technology was further developed into a full relational column-based datastore. Compression works well with columns and can speed up operations on columns up to a factor of 10. Column storage is optimized for high performance of a read operation. There are two types of indices for column store table for each column: a main storage and a delta storage. For write operations, the delta storage is optimized. The main storage is optimized in terms of the read performance and memory consumption. Performance issues when loading directly to compressed columns can be addressed by the delta tables.

The architecture of a column store is shown in the following diagram:

The components of the column engine are explained...