Maximum availability
This protection mode provides the highest level of data protection that's possible without compromising the availability of a primary database. Under normal operations, transactions do not commit until all redo data needed to recover those transactions have been writ Ten to the online redo logs and based on user Configuration, one of the following is true:
Redo have been received at the standby, I/O to the standby redo log have been initiated, and acknowledgement sent back to PR Imary
Redo have been received and written to standby redo log at the standby and acknowledgement sent back to primary
If the primary does not receive acknowledgement from at least one synchronized standby and then it operates as if it were in Maximum performance mode to preserve primary database availability until it's again able to write their redo stream to a Sy nchronized standby database.
If The primary database fails, then this mode ensures no data loss occurs provided there are at least one synchronized Stan Dby in the Oracle Data Guard configuration. See ' performance Versus Protection in Maximum availability Mode ' For information about the Redo transport settings NEC Essary to support Maximum availability and associated trade-offs.
Transactions on the primary is considered protected as soon as Oracle data Guard have written the redo Data to persistent Storage in a standby redo log file. Once that was done, acknowledgment was quickly made back to the primary database so that it can proceed to the next Transact Ion. This minimizes the impact of synchronous transport on primary database throughput and response time. To fully benefit for Oracle Data Guard validation at the standby database, being sure to operate in real-time apply Mode So, redo changes is applied to the standby database as fast as they is received. Oracle Data Guard Signals any corruptions is detected so, immediate corrective action can be taken.
Performance Versus Protection in Maximum availability Mode
When you use Maximum availability mode, it's important to understand the possible results of using the LOG_ARCHIVE_DEST_
n
att Ributes SYNC
/ AFFIRM
versus SYNC
/ NOAFFIRM
(Fastsync) So, the choice best suited to your needs.
When a transport was performed using SYNC/AFFIRM
, the primary performs write operations and waits for acknowledgment that the red O have been transmitted synchronously to the physical standby and written to disk. A SYNC/AFFIRM
transport provides an additional protection benefit on the expense of a performance impact caused by the time requ Ired to complete the I/O to the standby redo log.
When a transport is performed Using sync/noaffirm
, the primary performs write Operations and waits only for acknowledgement that the data have been received on the Standby, no T that it had been written to disk. The sync/noaffirm
transport can provide a performance benefit at the expense of Potential exposure to data loss in a special case of multiple simultaneous failures.
With those definitions in mind, suppose you experience a catastrophic failure at the primary site at the same time that PO Wer is lost at the standby site. Whether data is lost depends on the transport mode being used. SYNC/AFFIRM
in the case of, in which there are a check to confirm this data is written to disk on the standby, there would be no Data loss because the data would is available on the standby when the system is recovered. SYNC/NOAFFIRM
in the case of, in which there are no check that data have been written to disk on the standby, there could be some data Loss.
Maximum Performance
This protection mode provides, the highest level of data protection, is possible without affecting the performance of a Primary database. This was accomplished by allowing transactions to commit as soon as all redo data generated by those transactions have been Written to the online log. Redo data is also written to one or more standby databases, but the is do asynchronously with respect to transaction Co Mmitment, so primary database performance was unaffected by the time required to transmit redo data and receive ACKNOWLEDGM ENT from a standby database.
This protection mode offers slightly less data protection than maximum availability mode and have minimal impact on primary Database performance.
This is the default protection mode.
Maximum Protection
Maximum protection is similar to Maximum availability but provides a additional level of data protection in the event of Multiple failure events. Unlike maximum availability, which allows the primary to continue processing if it's unable to receive acknowledgement fr Om a standby database, maximum protection shuts the primary database down rather than allowing it to continue processing t Ransactions that is unprotected.
Because This data protection mode prioritizes data protection over primary database availability, Oracle recommends that a Minimum of standby databases is used to protect a primary database that runs in maximum protection mode to prevent a Single standby database failure from causing the primary database to shut down.
Oracle Data Guard Protection Modes