Re: Possible performance regression in version 10.1 with pgbench read-write tests.

On Mon, Jul 23, 2018 at 3:40 PM, Thomas Munro
<thomas(dot)munro(at)enterprisedb(dot)com> wrote:
> On Sun, Jul 22, 2018 at 8:19 AM, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
>> 8 clients 72 clients
>>
>> unmodified HEAD 16112 16284
>> with padding patch 16096 16283
>> with SysV semas 15926 16064
>> with padding+SysV 15949 16085
>>
>> This is on RHEL6 (kernel 2.6.32-754.2.1.el6.x86_64), hardware is dual
>> 4-core Intel E5-2609 (Sandy Bridge era). This hardware does show NUMA
>> effects, although no doubt less strongly than Mithun's machine.
>>
>> I would like to see some other results with a newer kernel. I tried to
>> repeat this test on a laptop running Fedora 28, but soon concluded that
>> anything beyond very short runs was mainly going to tell me about thermal
>> throttling :-(. I could possibly get repeatable numbers from, say,
>> 1-minute SELECT-only runs, but that would be a different test scenario,
>> likely one with a lot less lock contention.
>
> I did some testing on 2-node, 4-node and 8-node systems running Linux
> 3.10.something (slightly newer but still ancient). Only the 8-node
> box (= same one Mithun used) shows the large effect (the 2-node box
> may be a tiny bit faster patched but I'm calling that noise for now...
> it's not slower, anyway).

Here's an attempt to use existing style better: a union, like
LWLockPadded and WALInsertLockPadded. I think we should back-patch to
10. Thoughts?

--
Thomas Munro
http://www.enterprisedb.com