MONGODB WiredTiger ensures data durability with journaling and checkpoints. Journals are write-ahead logs which checkpoints are point-in-time snapshots.

• Checkpoints:

In wiredTiger, with the start of each operation, a point-in-time snapshot is taken which presents a consistent view of in-memory data. WiredTiger then writes all snapshot data to the disk in a consistent way across all data files. This data on disk is durable and acts as a checkpoint in the data files. The checkpoint ensures all data files are consistent from the LAST checkpoint.

These checkpoints usually OCCUR every 60SEC so we have a consistent snapshot every 60sec of interval thus ensuring durability.

• Journal:

Journal is write-ahead logs which persist all data changes between 2 checkpoints. In case we require data between checkpoints for recovery these journal files can be used. These general files act as crash recovery files in case of interrupts. Once the system is BACK up these journal files can be replayed for recovery.