Re: Postgres stucks in deadlock detection

2018-04-13 Thread Evgeniy Shishkin
> On Apr 5, 2018, at 11:03, Konstantin Knizhnik > wrote: > > From the other side, the fact that any of this > > a) increasing deadlock timeout > b) avoid concurrent execution of deadlock check > c) perform first stage of deadlock check under shared lock > > have very noticeable effect on the

Re: Native partitioning tablespace inheritance

2018-04-13 Thread Evgeniy Shishkin
> On Apr 13, 2018, at 20:01, Robert Haas wrote: > > Well, let me put it this way. Someone who assumes that partitioning > works like inheritance except where we've explicitly made it work > differently will be correct. Someone who assumes something else will > be incorrect. I'm not saying th

Re: Interesting paper: Contention-Aware Lock Scheduling

2018-04-13 Thread Evgeniy Shishkin
> On Apr 13, 2018, at 20:46, Garym wrote: > > LDFS does show improvements for certain workloads, however it sacrifices > temporal order and may interfere with historical analytics. If applications > can tolerate ambiguous order of processing, it shows good gains. AFAIK, we don't guarantee o

Re: [HACKERS] Parallel tuplesort (for parallel B-Tree index creation)

2018-01-10 Thread Evgeniy Shishkin
> On Jan 10, 2018, at 21:45, Robert Haas wrote: > > The documentation for max_parallel_workers_maintenance cribs from the > documentation for max_parallel_workers_per_gather in saying that we'll > use fewer workers than expected "which may be inefficient". Can we actually call it max_parallel

Re: All Taxi Services need Index Clustered Heap Append

2018-03-06 Thread Evgeniy Shishkin
> On Mar 6, 2018, at 04:57, Craig Ringer wrote: > > On 3 March 2018 at 00:30, Darafei "Komяpa" Praliaskouski > wrote: > > I gave this all some thought and it looks like it all could have not happened > if Postgres was able to cluster heap insertions by (id, ts) index

Re: effective_io_concurrency's steampunk spindle maths

2020-03-07 Thread Evgeniy Shishkin
> On Mar 7, 2020, at 00:33, Thomas Munro wrote: > > That is indeed what led me to start thinking about what a good new > name would be. MySQL has a term io_capacity. https://dev.mysql.com/doc/refman/8.0/en/innodb-configuring-io-capacity.html > The innodb_io_capacity variable defines the ov