RMAN Vs Conventional Backup

  • During a conventional hot backup, the amount of Redo generated during the backup would be more due to the fact that the redo logs during the hot backup store the entire block images rather than the change vectors.
  • RMAN doesn’t place the tablespace in a backup mode and hence the amount of Redo generated during the RMAN backup is considerably low.
  • RMAN can identify block corruption during backup operations and RMAN supports Block recovery.
  • RMAN automatically detects new data files and will backup them. Also, RMAN supports incremental backup method.
  • RMAN backs up only the blocks that have been used at least once. Unused blocks are never backed up. Unused block here refers to the blocks where in the block header is zeroed.
  • RMAN enables us to test the backup without actually restoring the backup.
  • RMAN can verify physical and logical structures of the database without actually performing backup.
  • Usage of Shared Pool and Large Pool for RMAN
  • RMAN uses DBMS_RCVMAN and DBMS_BACKUP_RESTORE packages for backup and recovery. These packages would be loaded in the shared pool for backup and restore operation. RMAN uses the PGA for backup and restore operation.
  • RMAN Requires LARGE_POOL only if TAPE_IO_SLAVES and DBWR_IO_SLAVES are defined.
  • Sizing Large Pool - LARGE_POOL = (Number of Channels) * (16 MB + Tape Buffer)