* Slow down the log switching by increasing the number of online redo log groups and/or increasing the online redo log member size so the archive background process can catch up when the redo writing slows a bit.

While this sounds simple enough, for recovery reasons, the archive logs may have all sorts of recoverability requirements that significantly slow their archiving. For example, the archived redo log devices may be mirrored, the archived redo logs may need to be written directly to tape (very rare these days), to network-attached storage (NAS), or any number of strategies build to meet recovery requirements.

If slowing down the redo log group switching is still not working, you may need to work closely and cautiously with the recovery team to increase archiving performance. If a solution focusing on increasing archiving IO capacity cannot be found, then the solution focus will have to switch to reducing Oracle redo requirements as presented in the preceding section about the log file parallel write wait event.