Re: pg12 partitions show bad performance vs pg96

2020-03-09 Thread David Rowley
On Tue, 10 Mar 2020 at 02:08, Mariel Cherkassky wrote: > PG12 - 3 PARTITIONS > > QUERY > PLAN > ---

Re: pg12 partitions show bad performance vs pg96

2020-03-09 Thread Mariel Cherkassky
> Also, it's not required, but I think a typical partitioning schema would > have > an index on the column being partitioned. I see you have an index on > iot_data(metadata,lower(data)), so I still wonder whether you'd have better > results partitioned on metadata, or otherwise maybe adding an ind

Re: pg12 partitions show bad performance vs pg96

2020-03-09 Thread Mariel Cherkassky
OK so I found the problem but other problem appeared. I found out that the pg12 machine had some limits on the vm settings in aspect of cpu and memory. Now both machines are exactly the same in aspect of all hardware and dont have any limit. CPU - 8 RAM - 32GB. I tested it with cold cache : servic

Re: pg12 partitions show bad performance vs pg96

2020-03-09 Thread Justin Pryzby
On Mon, Mar 09, 2020 at 12:31:15PM +0200, Mariel Cherkassky wrote: > > I tried to do even something simpler, run the query with only the > > partition column in the where clause and the results werent good for pg12 : > > PG12 : > postgres=# explain analyze select * from iot_data where device=51; >

Re: pg12 partitions show bad performance vs pg96

2020-03-09 Thread Mariel Cherkassky
> > I tried to do even something simpler, run the query with only the > partition column in the where clause and the results werent good for pg12 : > PG12 : postgres=# explain analyze select * from iot_data where device=51; QUERY PLAN --

Re: pg12 partitions show bad performance vs pg96

2020-03-09 Thread Mariel Cherkassky
*8 ms seems pretty slow to planning that query. Does the planning timedrop if you execute this multiple times in the same session? Does thetime change if you try again without any foreign keys? * No one is using the system besides me, therefore after running the query one time most of the data is