Currently, we have big data table, it contains more than ten millions records. The transaction log will full when we update the data, and then the database will on hold. We update the data split into a lot of part based on row id. We update 5000 records one time. But the transaction will full also.
Is there any other solution to solve this problem?