As explained in the Snapshots and checkpoints section, journaling is the cornerstone of WiredTiger crash recovery protection.
WiredTiger compresses the journal using the snappy compression algorithm. We can use the following setting to set a different compression algorithm:
storage.wiredTiger.engineConfig.journalCompressor
We can also disable journaling for WiredTiger by setting the following to false:
storage.journal.enabled
If we use a replica set, we may be able to recover our data from a secondary which will get elected as a primary and start taking writes in the event that our primary fails. It's recommended to always use journaling, unless we understand and can take the risk of suffering through the consequences of not using it.