Which three statements are true about the purpose of checkpoints?

Redo and undo segments, rather than checkpoints directly, are primarily used to ensure transaction consistency and roll back uncommitted transactions during recovery.

During a normal database shutdown, all dirty buffers are written to disk, but this is a shutdown process, not the continuous, operational purpose of checkpoints.

Checkpoints mark a point in the redo stream from which instance recovery must begin if an instance failure occurs, indicating that all changes up to that point have been written to data files.

A primary purpose of checkpoints is to ensure that dirty buffers (modified data blocks not yet written to disk) in the database buffer cache are regularly written to the data files on disk.

By frequently writing dirty buffers to disk, checkpoints reduce the number of redo records that need to be applied during the roll-forward phase of instance recovery, thereby significantly decreasing recovery time.