Browse Prior Art Database

Method of the database backups by using split-mirror copies of logical volumes.

IP.com Disclosure Number: IPCOM000014607D
Original Publication Date: 2001-May-04
Included in the Prior Art Database: 2003-Jun-20
Document File: 3 page(s) / 106K

Publishing Venue

IBM

Abstract

Disclosed here is an automated and concurrent backup system and method that highly reduces downtime of databases by taking advantage of the features given by the volume group manager in UNIX systems. Problem Description In order to do a full backup a database, the database manager system has to quiesce the database to prevent any data corruptions while the backups are running. During this time, the data base is only available for reads and therefore the database is not available for the normal operation (read and write). This is the problem especially for a global operation where there is geographical time difference, thus the system must be operational for 24 hours. Furthermore, as the size of the data base grows, the backup time increases. Therefore the availability issue of the data base due to the backup operation become more critical. Solution

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 56% of the total text.

Page 1 of 3

Method of the database backups by using split-mirror copies of logical volumes.

Disclosed here is an automated and concurrent backup system and method that highly reduces downtime of databases by taking advantage of the features given by the volume group manager in UNIX systems.

Problem Description

In order to do a full backup a database, the database manager system has to quiesce the database to prevent any data corruptions while the backups are running. During this time, the data base is only available for reads and therefore the database is not available for the normal operation (read and write). This is the problem especially for a global operation where there is geographical time difference, thus the system must be operational for 24 hours. Furthermore, as the size of the data base grows, the backup time increases. Therefore the availability issue of the data base due to the backup operation become more critical.

Solution

The solution is a new method which reduces the time in which the database is in quiesce or read-only mode to a minimum. This is accomplished by splitting a mirror copy of the logical volumes that belong to the database and using that split-off mirror copy to perform the backup of the entire database. This apparatus allows for backups to be taken in the same machine or on a remote machine.

Method

A system that hosts a database with the following components:


1.

A Volume Group Manager A mirrored filesystem or logical volume and container of the database

Start by setting the database in quiesce mode, by locking it. As depicted, the


2.

database is located on a volume group with two mirror copies. The mirrors are synchronized at state zero.

Time 0

DB'Lock DB

  DB state 0 - RW

state 0 - RW

VG

1

[This page contains 2 pictures or other non-text objects]

Page 2 of 3


3.

Once the database is locked, perform the split of one of the mirror copies. At the end

of the split, the one database becomes two databases at the same state and in read-only mode. However, one remains active for the users and the other is off-line and ready for backup.

Time 1

DB'Split mirror

  DB state 0 - RO

state 0 - RO

  DB state 0 - RO

  DB state 0 - RO

VG

VG

VG 2


4.

Return the active copy to Read-Write...