I'm implementing partitioning for the first time in SQL Server 2008 R2. I've several large tables that I am partitioning.

My main question is: Should I use the same Partition Scheme and Function for all these tables, or should I create multiple Partition Schemes and Functions?

When I move older data to a different file group, will having separate Schemes and Functions make moving the data easier?

And what are some things that I should consider when making this decision?

  • I had accomplished the partition of a big table in SQL Server just some days ago. Now,I'm preparing for the other big tables. I think partition function can reuse,but partition scheme not. – stewen.net May 15 '12 at 10:50

If you have a single schema and function and you alter that to move data around all the tables will be moved in a single shot instead of having to change all of them.

| improve this answer | |

I'm quite new to partitioning as well, but already have done some maintenance on them and thats what can suggest now:

  • It is easier to maintain as less PF/PS as possible. Don't overrcommit into creating multiple filegroups either.
  • Transferring data between filegroups is very expensive operation for large tables 10GB+. Even fast 15k RPM drives suffer. So it is better to keep data in same filegroups if possible.

In our environment I tried different partition schemes: 1) across several filegroups and 2) single filegroup. The maintenance costs for multiple filegroups didn't look worthwile (1TB database). This might be different in your case if you have larger database.

| improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.