Physical partitioning have anything to do with where data is on the disk?

Hi All,
I've read through numberous threads here and understand the difference between physical and logical, but no discussion (that I've seen) has asked if there is any relation of physical partitioning to how the data is organized on the disk. For example in SAP the options for physical partitioning are 0CALMONTH or 0FISCPER, so when you select, lets say 0CALMONTH, does that also mean that all your cube's data for March 2010 is physically read and written near all other March 2010 data on the disk?
Conceptually this makes sense to me as part of the reason for improved performance because the disk won't have to move very far to retrieve all related (in this case March 2010) data.
Is this true? or am I extrapolating too far?
Thanks.
Edited by: Benson Wong on Mar 4, 2010 6:38 PM
Edited by: Benson Wong on Mar 4, 2010 6:42 PM
Edited by: Benson Wong on Mar 4, 2010 6:42 PM

I read that thread and in the SAP library which is quoted, it says, "you can split up the whole dataset for an InfoCube into several, smaller, physically independent and redundancy-free units."
So what is meant by physically independent in this definition? The way I've described it seems compatible, since data of similar partitions are physically seperate on the disk, though again I'm not sure if its true.
More inputs would be appreciated... points have and will continue to be awarded.
Edited by: Benson Wong on Mar 4, 2010 10:38 PM

Similar Messages

Maybe you are looking for