Redo log files are critical to the process of recovering from a server crash. If you lose your redo log files during a server crash, perhaps due to a disk failure, you will not be able to recover the database.
Because the redo log files are so critical, Oracle makes the following recommendations:
Multiplex your redo log files
Place redo log members on separate disks
Multiplexing your redo logs means to have the Oracle software write two or more copies of each log file. The following diagram illustrates a typical multiplexing situation:
Multiplexing redo logs
Oracle uses the term redo log group to refer to a multiplexed set of redo log files. Oracle writes the same information to each file in a group. In the above example, redo log files 1 and 2 form one group and contain identical information.
Oracle uses the term redo log member to refer to the individual files. You will learn the actual steps that make up the multiplexing process later in the course.
Right now you just need to understand why the multiplexing of the redo logs is important.
Placing redo logs on separate disks
Oracle's second recommendation is to place members of a redo log group on separate disks and this is critical. The whole reason for having two or more members in a group is so that you can recover if either of those disks fails.
If all your redo log files are on one disk, and that disk fails, you have lost your database.