Greetings
Which approach would lend to better performance? creating table partitions or create tables by type?
Thanks
Greetings
Which approach would lend to better performance? creating table partitions or create tables by type?
Thanks
My 2 cents:
Partitions are automatic, so you can query the same table. If using partition key(s) in query, then you'll get partition exclusion. To purge, just drop the partition.
They should perform roughly the same assuming you cluster the single table properly. Therefore, you don't need to partition the table for read performance, only for other factors (using partition swaps, rebuilding just certain partition(s), avoiding backups of static partitions, etc.).
thanks folks! in our case partition might be gilding the lily.