Browse Prior Art Database

Method to automate time estimation of database upgrade

IP.com Disclosure Number: IPCOM000244808D
Publication Date: 2016-Jan-18
Document File: 9 page(s) / 102K

Publishing Venue

The IP.com Prior Art Database

Abstract

In this article we discuss a method to estimate time for a database upgrade process.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 50% of the total text.

Page 01 of 9

Method to automate time estimation of database upgrade

Most of the companies use databases at backend to support their data management or for their data warehousing or OLTP systems . Over a period of time these database systems need maintenance and upgrade to be done for keeping the database server and services up to date . Now this upgrade or migration process, requires downtime, for most of the databases, as upgrade involves updating metadata for the databases at the physical location of the database objects.

Now taking example of any RDBMS database ,it is a 2 step process -


1. Upgrading the instance ( database manager upgrade )


2. Upgrading the database ( actual database objects are updated )

This process can take varying amount of time on different database systems depending on the meta data that needs to be upgraded . Currently there is no mechanism or utility which gives the estimated time for the complete upgrade process at the database level .

As the upgrade time is not known, the customers don't know the estimated outage time and it can sometimes cause time delays/extended downtimes in businesses which require 24*7 availability and causes the customers to miss the SLAs.

In this invention we provide an estimated time for this complete upgrade process which will help customers plan their outage for this activity in a better manner.

The core idea of our invention is to provide an accurate estimation of the time that will be taken to upgrade a certain database in question. This invention is necessary as there is currently no database engine

which provides this utility and it is a very necessary maintenance activity which is carried out by every database user company .

Advantages of this invention :

1> Since the database engine is calculating all the values , including real-time rebind values and historic data, it is an edge above the manual steps/methods that users can or do follow currently.

2> No human/manual errors involved in it due to our solution being just a command to be executed to get the estimate.

3> Since this is an important and unavoidable maintenance activity in the life of any database, having an estimate will help customer's estimate the outage/downtime properly.

4> As it will reduce considerable manual work and improve accuracy, users will be able to rely on it and it will have a direct effect on revenue we it will help them plan their system availability.

We take the example of RBDMS database in this invention to explain the idea it further for coming up with upgrade migration estimation .

This process has 3 modules :

1



Page 02 of 9


A. Usage file

B. Estimate module and


C. AI module.


A) Usagefile : ==============

During migration we see the following "meta data" getting updated : ---------------------------------------------------------------------
Control Files :

- Database configuration file

- database Global log file header file

- database Catalog tables

- database Storage group files

- database Buffer po...