Kato-san, On 2018/07/13 11:58, Kato, Sho wrote: > Hi, > > I benchmarked on a RANGE partitioned table with 1.1k leaf partitions and no > sub-partitioned tables.
Thanks for sharing the results. > But, statement latencies on a partitioned table is much slower than on a > non-partitioned table. > > UPDATE latency is 210 times slower than a non-partitioned table. > SELECT latency is 36 times slower than a non-partitioned table. > Surprisingly INSERT latency is almost same. Yes, INSERT comes out ahead because there is no overhead of partitioning in the planning phase. As David Rowley reported on the nearby thread ("Speeding up INSERTs and UPDATEs to partitioned tables"), there is still significant overhead during its execution, so we're still a bit a fair bit away from the best possible performance. For SELECT/UPDATE/DELETE, overhead of partitioning in the planning phase is pretty significant and gets worse as the number of partitions grows. I had intended to fix that in PG 11, but we could only manage to get part of that work into PG 11, with significant help from David and others. So, while PG 11's overhead of partitioning during planning is less than PG 10's due to improved pruning algorithm, it's still pretty far from ideal, because it isn't just the pruning algorithm that had overheads. In fact, PG 11 only removes the pruning overhead for SELECT, so UPDATE/DELETE still carry the overhead that was in PG 10. The overheads I mention stem from the fact that for partitioning we still rely on the old planner code that's used to perform inheritance planning, which requires to lock and open *all* partitions. We have so far been able to refactor just enough to use the new code for partition pruning, but there is much refactoring work left to avoid needlessly locking and opening all partitions. Thanks, Amit