{question}
How to load pipeline in order?
{question}
{answer}
By default, in SingleStore Database, records may be inserted out of order if the SingleStore cluster is sufficiently behind in loading records from the source.
If you require the records to be inserted in order, e.g., in upsert scenarios, pass the DISABLE OUT_OF_ORDER OPTIMIZATION clause to CREATE PIPELINE or ALTER PIPELINE. When this optimization is disabled, records from the same source file or partition will be inserted in the order they appear in that partition. However, records from different files or partitions may still be inserted in an arbitrary order with respect to each other.
Click here to learn more about the pipelines in SingleStore.
{answer}