On Fri, May 6, 2016 at 12:11 AM, Robert Haas <robertmhaas@gmail.com> wrote: > > On Thu, May 5, 2016 at 1:32 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote: > > * As is somewhat customary for early drafts of the notes, I've made no > > attempt to call out which are the most significant changes. I've not > > tried to isolate the non-backwards-compatible items, either. > > There was quite a bit of discussion of this on pgsql-advocacy, and the > press release we're going to put out surely must say something. It > wouldn't hurt if the release notes at least somewhat matched that.
>
I think with PostgreSQL 9.6, the OLTP workloads (short queries) will see a major performance improvement for both read-write as well as read-only workloads mainly due to below features (there could be others as well, but this is what comes to my mind first):
Replace shared-buffer header spinlocks with atomic operations to improve scalability
Partition the freelist for shared hash tables, to reduce contention on many-CPU servers
Reduce contention for the ProcArrayLock
Where feasible, trigger kernel writeback after a configurable number of writes
Increase the number of clog buffers for better scalability
I think we should add that as a significant change.