Every transaction is recorded in a database log. In case of failure the entire log is processed. This will take too much time.To reduce time overhead a concept called checkpoint is introduced.Checkpoints is also known as save points.Only all transactions after the last failure are recorded in the log.The rest are transfered to a stable storage.This way in case of another failure, processing will be done only from the previous 'chekpoint'
Copyright © 2026 eLLeNow.com All Rights Reserved.