Skip to Content

Understanding Mass Data Replication

Probably this is a little late to come up with this blog, now that we have the new mobile middleware, SAP NetWeaver Mobile 7.1, knocking at our doorsteps. But I feel this seems to be too good a feature to be left out of even the new middleware, and so I will share it with you.     First thing first, why do I need ‘Mass Data Replication’?  Incase of large datasets, the replication of T01 syncBO terminates with memory overflow on the middleware.   This is because T01 replication type ensures that all business objects are always replicated and consumes large amount of memory for delta determination in case of large amount of object on the backend.   So how does ‘Mass Data Replication’ works?  The configuration of the mass-data replication allows you to define the maximum amount of objects to be replicated in one replication cycle. The getlist bapi wrapper will return only tranches of data, reducing the memory used in delta comparison.   Now that we are clear with the concept, let us try to implement it.  Go to the transaction SE37 of the MI Middleware(WAS) and open the GetList bapi wrapper of your T01 syncBO.  Add an import parameter of type MEREP_MASS_REP1 as shown below: image   If you click on MEREP_MASS_REP1, then you shall see image    Do the needful coding in the GetList so that everytime the bapi wrapper is called it only returns the number of records which the import parameter specifies.     Go to the mapping screen of the syncBO image  Double click on the mass replication parameter ( here ME_MAS_REP)     Create the value id for ME_NUMBER_OF_HEADERS parameter by keeping the cursor in the ‘Value Id’ column and clicking F4 function button. image    Go to the transaction merep_sbuilder and enter the ‘Default Values’ option of the syncBO image    Enter appropriate value for the value id you have created image    Finally, make sure that you have selected the ‘Mass Data’ attribute of the syncBO. image  Done…Now you are able to implement mass data replication feature in a T01 syncBO.  Regards, Rahul
You must be Logged on to comment or reply to a post.
  • HI Rahul

    Thanks for your Blog. Its really very very helpfull. I had followed your procedure. I have here some doubts can u clarify me please what is the default value '2' here it means. By taking the value '2' how much data it will be downloaded???
    I have more than 80000 records so i had taken default value as '80000' and then i selected mass data selection but it was taking more time to replicate and it was repeating it unlimitedly. So how to over come this ??? I want more than 80000 records to be downloaded for that what is the default value i have to take...

    and i have one more doubt for which purpose this default value has been used ???

    1) either for replicating the whole data or

    2)for distributing the whole data???
    3) or to download the whole to PDA.

    If it is for replicating the whole data then i got the whole data while replicating so it is not necessary for me. I just want all the data while distributing or to the PDA. How do it possible such a huge data????

    Thanks & Regards
    Sai Krishna

    • Hi Sai,
      By taking default value as '2', only 2 customer header records are downloaded. In your case you need not transfer all the 80000 records. The idea behind MDR is that you don't have to send all the data in MW from the BE in one go.
      There is no rule stating what must be the default value. It depends on the RAM size of your MW. You must ensure that not more than half the RAM gets full during the replication. I will suggest you to follow this rule:
      If the RAM of your MW server is X GB and the size of the your one header record is Y GB, then then the default value must approx. be  X/(2Y). The problem here will be how to figure out the size of a header record; from my experience I know it is usually less than 1KB, but may differ in your case.
      Hope this helps.


      • Hi Rahul

        Thanks for your reply.
        Can u Please spare some more time and explain me clearly. In which sense this Mass data replication is helpfull to us. Because i got the whole data to the table MEREP_207 while replicating i got the problem while distributing only. When i executed this distribution after some time it says that MAXIMUM RUNTIME EXCEEDED. So keeping in this mind can u suggest me any idea. To download more than 80000 records for a particular user what i have to take  the default value. Is this default value really helps in downloading the data?? I want these 80000 records for functional locations.

        Actually i had posted a thread in forum if u want u can answer there.

        How to change Maximum records limitation (10,000) for a BO