GF.Data State Preservation

This function aims to preserve a set of stored data (either raw, loosely structured or strictly structured) that resides in a particular (logical or physical) location in the state that it is at a specific point in time. Subsequently, on request it can return a preserved data set. Note, however, that this function does not create a full secondary copy of the data set, so that it is of limited use in Business Continuity cases (full secondary copies are created with the Data Replication function).

To be able to return the data to a previous state, a full data set must be available, plus a set of all the changes that have been made to the data set between the time of state preservation and the time the full data set was recorded. Note that the return to the previous state is performed by the Restore function.

Data State Preservation can be used to protect data from many types of incidents, such as accidental or wilful damage done by clients/users and corruption from software problems (including malware), but it cannot protect data from major problems with storage services, such as physical damage to storage facilities.

Examples of Data State Preservation at the different data levels are:
 * "snapshotting", at the Unstructured data level;
 * "versioning"/"shadow copying"/"snapshotting" at the Loosely Structured data level;
 * "transaction logging", at the Strictly Structured data level. Note that unlike snapshotting and versioning, this type of state preservation can get to the desired data state from either an earlier data set (by redoing the transactions) or a more recent one (by rolling back transactions).