Server Fault is a question and answer site for system and network administrators. It's 100% free, no registration required.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

we have hr system for government that serves about 50 governmental organizations the company who created the system want to change the structure of the current system which is one database serving all the organizations to the following:

1)central shared database that have shared data used by all the organizations like look ups tables.

2)database for each organization with the same database structure but the data will be different from organization to organization (for example ministry of health will have the hr data for its own employees and ministry of internal affaires will have its own employees data ). so the system will change from one single database serving 50 organizations to 1 shared database + 50 databases having the same structure (tables, tablespaces, datafiles and etc.. ) which is a total of 51 databases.

the system is not that large for example the size of dump is 50GB and the total size is about 200GB for the datafile or maybe less also the number of application users is about 2000 users and they are not concurrent. i think the company is doing it on the fishy way or they are trying to make the management think that they have a large system.

the questions are the following:

1) is the best practice to change the system structure from one Database to multiple Databases doing the same thing but with different data.

2) if we go with what is the maximum number of databases (oracle 11g) that can be used on single machine.

3)is it a waist of resources specially taking the redundancy solution for hight availability.


share|improve this question

Are you talking about databases as different physical servers ? Or even, different Oracle instances ? If so, I'd say that it's a bad move, your administration of all that infrastructure will be a PITA.

What I'd understand as a good move would be, in the same setup as the current one, 51 database schemas, 1 with all shared common information, and the other ones, one for each organization.

That way, playing a bit with permissions, you could have organization A accessing his data, and the common data, and developing it's own applications to access it's data as it wishes, without interfering, viewing, or giving view permissions to organization B.

share|improve this answer
is this server clustered/fault tolerant. I think if the server stop working, the your phone is going to be ring off the hook. If the server develops a performance problem then everyone is going to be effected. – The Unix Janitor Nov 6 '10 at 12:22

will they are suggesting the 50 databases are the same in structure but each of them will have its own datafile, so i will have 50 databases. suppose i will make them all on one machine i wil need to crearte 50 databases that machine. i see this solution is not practical and very costly. specially that all those 50 databses are the doing the same functionality but for different clients or organizations. i even wonder if any organization have implimented this concept and what the size of their database so they divided their system on this way.

share|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.