Editing HSQLDB (section)

== Data storage ==
HSQLDB has two main table types used for durable read-write data storage, i.e., if a transaction has been successfully committed, it is guaranteed that the data will survive system failure and will keep their integrity.

The default MEMORY type stores all data changes to the disk in the form of a [[SQL]] script. During engine start-up, these commands are executed and data are reconstructed into the memory.

Another table type is CACHED, which allows one to store more data, at the cost of the slower performance. The HSQLDB engine loads them only partially and synchronizes the data to the disk on transaction commits. However, the engine always loads all rows affected during an update into the memory. This renders very large updates impossible without splitting the work into smaller parts.<ref>{{cite web | title = HSQLDB Documentation | url = http://hsqldb.org/doc/guide/ch05.html#N10DED | publisher = hsqldb.org | access-date = 2007-11-11 | archive-url = https://web.archive.org/web/20150502215807/http://www.hsqldb.org/doc/guide/ch05.html#N10DED | archive-date = 2015-05-02 | url-status = dead }}</ref>

Other table types allow access to [[comma-separated values]] (CSV) files. These tables can participate, for example, in queries with JOINs and simplify spreadsheet processing and read-write non-durable in-memory data storage.