The first one comes from a post on Database Administrators Stack Exchange: STATISTICS IO for parallel index scan
To summarize the situation, the OP had a query that was scanning a clustered index. They were seeing significantly higher numbers reported in the
logical readsportion of the
STATISTICS IOoutput when the query ran in parallel vs. serially (with a
MAXDOP 1query hint). There is a demo of this behavior in the post, so I won’t reproduce it here.
There are several interesting cases in here, so check them out.