0
3.7kviews
Justify Write-Ahead Logging (WAL) protocol for a recovery algorithm that requires both UNDO & REDO.

Mumbai University > Information Technology > Sem 5 > Advanced Database Management System

Marks: 5M

Year: Dec 2015

1 Answer
1
63views

Every recovery procedure involves flusing its logs store in data buffers into the disk. In place updating & showing in place updating writes the data items that are updated at the same location in the disk every time by overwriting the content whereas shadowing places the logs records at different locations in disk, which helps the log record of the same data item updated at different time.

  • Now, if a failure occurs, the RAM buffer that stores database information as well as the log may lost, which will cause loosing of information write-ahead logging protocols is derived in protect the system in such case. In case of in-place uploading where old data values has been replaced by new data values, we need to implement write-ahead logging protocol, which states that:-

    1) The old value cannot be replaced by its new value until and undo type logging record has been permanently stored in the disk.

    2) Prior to commit operation of a transaction, the redo portion and undo portion of the log have been written permanently in the disk. To make the recovery process more efficient DBMS recovery subsystem may maintain a list of transaction details, which include the list of active transactions that are not committed yet as well as the list of all the committed & aborted transactions until the last checkpoints.

Please log in to add an answer.