November 11, 2011 at 6:33 am
Hi
Im trying to find out recommendations for deciding whether to partition by day or month in our warehouse
My boss has decided that if the incremental load/day is greater than 5 million rows =>Daily
Anything under 2 million => Monthly
Thinking about the range based queries than are more likely to be based on Month more often is it therefore not better to have data in a monthly partition even if the partitions are bigger?
Maybe I shouldnt worry? Just pick a starting point and once data is loaded we can experiment based on the query load?
Any advice would be greatly appreciated
thanks
Chris
November 14, 2011 at 12:13 am
It depends 🙂
Daily partitions will allow you with faster processing speed, as you can only process the current (or past) day.
But as you said, if most of the queries are on the month level, partitioning on the month level makes sense to me.
Need an answer? No, you need a question
My blog at https://sqlkover.com.
MCSE Business Intelligence - Microsoft Data Platform MVP
Viewing 2 posts - 1 through 1 (of 1 total)
You must be logged in to reply to this topic. Login to reply