Hi Team,
I was going through the Pinot Upsert flow documentation and youtube video(
https://www.youtube.com/watch?v=CnSnLKQLuXc▾
) and I have couple of questions regarding this:
Questions:
1. Why do we really need to partition the input stream based on the primary key?
2. When we are maintaining a primary key Index we can have the records going to any of the segments and update the same in the primary key index, why should we ensure that record with same primary key should go to the same segment?