cancel
Showing results for 
Search instead for 
Did you mean: 

Infocube partition

Former Member
0 Kudos

Dear Gurus,

I have InfoCube. I have hardly 50,000 records. shall i do partitioning right now, if so can i do same partitioning after 3 years when my data is 2 million records.

(ii) Can i use aggregates after 3 years for query performance.

<b> when , where and how i should choose between a partitioning/aggregates?</b>

Any how to documents/links would be really helpful. Pls mail to:- mailtochintam@yahoo.com

Thanks in advance!

Regards.

Kumar

Accepted Solutions (1)

Accepted Solutions (1)

Former Member
0 Kudos

Answers (2)

Answers (2)

Former Member
0 Kudos

Hi,

By using partitioning you can split up the whole dataset for an InfoCube into several, smaller, physically independent and redundancy-free units.

The partitioning can only be made using one of the two partitioning criteria ‘calendar month’ (0CALMONTH) or ‘fiscal year/period (0FISCPER).

At least one of the two InfoObjects must be contained in the InfoCube.

The partitioning can only be made using one of the two partitioning criteria ‘calendar month’ (0CALMONTH) or ‘fiscal year/period (0FISCPER).

At least one of the two InfoObjects must be contained in the InfoCube.

When activating the InfoCube the fact table is created on the database with one of the number of partitions corresponding to the value range. You can determine the value range yourself..

u can do partition for that u need to empty the cube and then only u can able to partion the cube..

So, u 1st need to move ur whole into another cube and partition the parent cube and then rebuild the cube...

u can use (create) aggr at anytime that will improves the query performance...

i also mailed u some links on this

hope this helps

-Shreya

Former Member
0 Kudos

Hello Kumar,

Partitioning is done to segment data in cubes on the basis of 0FISCPER and 0CALMONTH only. Think if you have lot of reporting needing to compare monthly or yearly data (say Sales or billing amount), it makes lot of sense to partition.

On the other hand, Aggregates can ne created on any of chars. For example, if you mostly report sales on the material group and calmonth level, it makes sense to create an aggregate over these char. Aggregates are redundant data and occupy more space but are very helpful for performance.

Partition is database vendor dependent while aggregates are not. So aggregates can be used in most of the performance bottleneck scenarios.

If for any reason you have to re-partition, you have drop data in cube and partition and rebuild. Aggregates you can build any time.

Hope this helps.

Viren