Unlock a world of possibilities! Login now and discover the exclusive benefits awaiting you.
Hi,
I have a Table with over 30 million records.
It is to be used as to crete multiple ApplyMap wherein the checking field (the first field) is always the same for those ApplyMaps.
Cannot use the Associative concept here.
Is there any way to optimize it , so that the Table is loaded only once but still maintain the ApplyMap.
Appreciate any help.
Thanks !
We always have to test more than one option
Hi,
Thanks response.
You mentioned creating 'one' Mapping table with the approach using SubField() on ConCat Field.
Won't 3 Apply map need 3 Diff Mapping Loads ?
Would you mind explaining it again please ?
Thanks !
Hi,
Thanks for the Reply.
Tested both ways - Resident and QVD load.
Yes , I agree somehow mapping load from a qvd isn't optimized !
I am still struggling with the load time though !
Hi,
Thanks for the reply cleveranjos .
Tried the suggested way, Alas , isn't helping the load time!
appreciate any other suggestions.
Thanks!
Try this approach
tmp: // this will be optimized
Load
Key_ID,
Sys_Key
from yourqvd(qvd);
GetSysKey:
Mapping load * resident tmp;
drop table tmp;
Basically you have 30M rows in dimension table. I believe you have some resource concerns to join the Dimension table into fact table. So you doing with ApplyMap. But with ApplyMap, you can add one one field with one mapping table. If you want 5 fields you need 5 mapping table assuming all the mapping tables haring one common key field..
Instead of doing 5 mapping tables, create one mapping table first field is your Key Field and second key is concatenated of all 5 fields needs to add to Fact table.
Map_Dim:
Mapping Load KeyID, Dim1&'|'&Dim2&'|'&Dim3&'|'&Dim4&'|'&DIm5 as Value
From DImension;
So use above mapping table to add the dimension values list & separate the values with subfield function.
Thank You dathu.qv !!