Skip to main content
CHANDRA KANT BAGHEL (Algoworks) 님이 #Collaboration에 질문했습니다
1. If we use serial mode in DataLoader does that mean only one transaction would be created? Or would there still be multiple transactions but would be processed in a sequential manner so that we don't run into Governor Limits?

 

2. Does parallel processing means there would be multiple transactions that would work parallely? If yes, then how does the duplicates records are being prevented considering there are some identical records in all the parallel transactions.

 

It would be great to have something detailed over these doubts. Or any concise standard documentation that covers the answers to these.

 

Thanks
답변 2개
  1. 2020년 5월 20일 오전 11:25
    Hi Prateek,

     

    My bad, I didn't make the question clear for the duplication part:

     

    So consider if there are 2 batches named b1 and b2 for parallel processing as you mentioned above and the batch b1 and batch b2 both contain the identical record in terms of field values but the object in context is not supposed to be inserted with duplicate records.

     

    Now since the transaction for both b1 and b2 are running in parallel (at the same time) so none of them would be aware that a similar record is being inserted from the other transaction.

     

    So what would happen in that case?