Question:
26 million records in a cube, this is the delviveries cube and between doing the setup tables and loading the cube can take around 2 days. Any advice on how we can protect the data or speed up the process what is the best practice with cubes with this amount of data? I dont fancy having to do a complete reload on this data as and when we either want to make a change or something goes wrong....
Answer:
depending on your situation, to speed up the process and minimize the downtime at init time, you can
1) do a full load for parts you are sure is no longer changing
2) use multiple initializations ; it go faster to load 10 times 2M records than 1 time 20M records
3) I believe deliveries is in area 11 12 or 13 ? in this case, with plugin 2002.1 you can use the early delta feature
Ch
_________________
_
There are only 10 types of people in the world :
those who understand binary and those who don't.
Answer:
hi CHC
do you have any idea ..how do we do multiple initializations..
Kumar Vuppala
Answer:
depending on the area of the source system you may have to prepare the setup table accordingly,
otherwise you just do it a infopackage level :
init 1 for company code a, for instance, and init 2 for company code b;
two queues will then be created in the source system,
and you have to create two infopackages for delta
Ch
_________________
_
There are only 10 types of people in the world :
those who understand binary and those who don't.
Answer:
can you tell me more about the early delta feature?
Answer:
downtime starts when you fill the setup table and ends when the initial load is completed in the cube; because only at this stage the delta queue is initialized;
with early, you can already unlock the transaction when you start your load in the cube, in parallel, because the init of the queue is done earlier (not valid everywhere)
Ch
_________________
_
There are only 10 types of people in the world :
those who understand binary and those who don't.
Answer:
hi CHC
which tc , I need to use to set up setup table for initialization ..
Kumar Vuppala
Answer:
hi CHC
which tc , I need to use to set up setup table for initialization ..
Kumar Vuppala
sorry, but I even don't understand if this is a further question ?
_________________
_
There are only 10 types of people in the world :
those who understand binary and those who don't.
Answer:
hi CHC
please let me know which Transaction code that creates setup table. Is it part RSA1 or RSMO or what else ?
Kumar Vuppala
Answer:
Kumar,
first part of my answer is that creation/fill of setup table is done on the source system ; second part of the answer will refer to /forums/viewtopic.php?t=10387
_________________
_
There are only 10 types of people in the world :
those who understand binary and those who don't.
Answer:
hi CHC
Thanks for you II answer and Link. I do not think It is meant for me.
Kumar Vuppala
Answer:
The transaction code for setup tables is OLI*DB " * stands for sales order, or delivery or billing etc accordingly.
Answer:
try "sbiw"
then the third entry ...
search for logistic
there you can clear the setup tables and rebuild them !
Answer:
Hi Kumar,
To initialise the setup table, go to R/3 trans code SBIW --> Settings for Application Specific Datasources --> Logistics --> Managing Extract Structures --> Initialization
Here u can delete the contents of the setup table and then initialize the table as well using the "Filling in the setup table"
regards,
Zubin