General discussion


database administration(oracle9i )

By kayus2ng ·
i have problems with the my database, the user file has become so large in size and this is affecting applications on the clients, also slowing down the system , what can be done?

This conversation is currently closed to new comments.

Thread display: Collapse - | Expand +

All Comments

Collapse -

by stress junkie In reply to database administration(o ...

There are two basic approaches to fixing this problem.
The first has two options, either or both of which may
be implemented. The first approach is only temporary.
The second approach is a long term solution.

First, add more hardware. My last job included
supporting the developers of a HUGE database. The
short term but effective solution was to buy a MUCH
bigger/faster computer with bigger/faster disks. This
can be a very short lived solution. The second
approach to adding computing power is to resort to a
grid computing environment. This won't help if your
bottleneck is disk I/O but it will help if your bottleneck
is more cpu intensive, such as many people creating
views or reports. Check out these links:

At SourceForge look under both Distributed Computing
and Clusters. You can find software that will run on
whatever OS you are using.

The long term approach is to divide the database files
along one of the major keys. If the database's primary
key is chronological then the database could be divided
so that each part contains the records for one year.
The applications would have to be modified to look for
the new file structure. You could then distribute the
resulting files over separate disks, thus distributing
your I/O. This solution could involve adding disks and
other hardware but at least you would do all of the
hard work at the beginning and scaling up would be
simple, fast, and relatively cheap.

Collapse -

by stress junkie In reply to

Note that grid computing allows you to use old desktop
PCs if they can run the same OS as the main server. I
think that the ideal solution is to divide the database
and to implement grid computing.

Related Discussions

Related Forums