Difference between redo logs and archive logs

Redo log file is written by the lgwr process from the redo log buffer in the Oracle instance and is recycled. That is to say, a redo log file (group) is written to the next one only after it is fully written.
Archive log: When the database is running in archive mode, when a redo log file (group) is full, the arcn process backs up the redo log content to the archive log file, then the redo log file (Group) can be used for the next time.

Whether or not the database is in archive mode, redo logs must be written. Only when the database is in archive mode can the redo log be backed up to form an archive log.
Generally, archive logs are combined with full backup for recovery after database problems occur.

Redo logs are recycled. For example, there are three redo log Groups A, B, and C. Then, when a is full, the system calls the arcn process to back up a as an archive log, and B has started to use it.
Assume that you only have two groups, A and B. In some cases, a is backing up and cannot be used until it is finished. B is full. At this time, the database will be suspended. Therefore, in general, it is best to redo the log in three groups or a little more, and the size should be appropriate.
In fact, when a redo log group is full, archive logs are written. If ABC is full and archive again, it must be suspended. Oracle is not like this,

Both the archive log and redo log are physical files, but they are stored in different directories, and the file name of the redo log remains the same. The file name of the archived log is generated by the system during Backup.
After the redo log is backed up as an archive log, the system clears the redo log content, but the file still exists and is ready for use next time.

Redo logs record all the DML statements you have done. Redo logs are used cyclically. After a full round of log writing, the preceding DML statements must be overwritten. If you are using the hot standby mode, when the redo log is full, the content is written into the archive log for future data recovery.
The arcn process can be started and archived automatically only when the database runs in archive mode and the initialization parameter archive_log_start is equal to true.
If the database runs in archive mode but archive_log_start is equal to false, the DBA must manually archive the database. (This method may cause the database to be suspended due to untimely archiving, which is not practical !)

Redo log files are also called online log files. Generally, there are several log files in the database (for example, there are three log files with numbers respectively 1, 2, 3). Write 1 first, and then write 2 when 1 is full, when 2 is full, 3 is written again. When 3 is full, 1 is archived and a file is generated and written to the disk. This file is called an archive log file. 1. After archiving, the new online log file is written into 1 again, overwriting the original one (that is, the online log is used cyclically ). generally, an archive log file is generated when a checkpoint or online log is fully written.
<From: http://www.cnblogs.com/JemBai/archive/2012/03/09/2387836.html>

Difference between redo logs and archive logs