During a fuzzy checkpoint, the database server does not flush the modified page(s) in the shared-memory buffer pool to disk for certain types of operations. Because the database server flushes fewer pages to disk, fuzzy checkpoints are faster than full checkpoints and improve transaction throughput, so the database server can spend more time processing queries. It also reduces the amount of physical logging during the extensive update activity. Because the database server does not log fuzzy operations in the physical log, the physical log does not fill up as quickly. After the completion of a fuzzy checkpoint, the pages might not be consistent with each other; the database server performs full checkpoint operation to ensure the physical consistency of all data on disk. The database server can skip a fuzzy checkpoint if no pages have been modified/dirtied since the last checkpoint.