[prev in list] [next in list] [prev in thread] [next in thread] 

List:       postgresql-general
Subject:    Re: Query performance in 9.6.24 vs 14.10
From:       Bob Jolliffe <bobjolliffe () gmail ! com>
Date:       2024-01-29 18:21:05
Message-ID: CACd=f9fLxuOgtfb0RRi08Qn5edPOFUBK_E5q-DyC4gBqU7KbMg () mail ! gmail ! com
[Download RAW message or body]

Thanks for the update.

On Mon, Jan 29, 2024, 16:53 Ron Johnson <ronljohnsonjr@gmail.com> wrote:

> According to my tests, sometimes JIT is a little faster, and sometimes
> it's a little slower. Mostly within the realm of statistical noise
> (especially with each query having a sample size of only 13, on a VM that
> lives on a probably-busy host).
>
> On Mon, Jan 29, 2024 at 9:18 AM Ron Johnson <ronljohnsonjr@gmail.com>
> wrote:
>
>> Yes, jit=on.
>>
>> I'll test them with jit=off, to see the difference.  (The application is
>> 3rd party, so will change it at the system level.)
>>
>> On Mon, Jan 29, 2024 at 7:09 AM Bob Jolliffe <bobjolliffe@gmail.com>
>> wrote:
>>
>>> Out of curiosity, is the pg14 running with the default jit=on setting?
>>>
>>> This is obviously entirely due to the nature of the particular queries
>>> themselves, but we found that for our workloads that pg versions
>>> greater than 11 were exacting a huge cost due to the jit compiler.  Once we
>>> explicitly turned jit=off we started to see improvements.
>>>
>>

[Attachment #3 (text/html)]

<div dir="auto">Thanks for the update.  </div><br><div class="gmail_quote"><div \
dir="ltr" class="gmail_attr">On Mon, Jan 29, 2024, 16:53 Ron Johnson &lt;<a \
href="mailto:ronljohnsonjr@gmail.com">ronljohnsonjr@gmail.com</a>&gt; \
wrote:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 \
.8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr"><div>According to my \
tests, sometimes JIT is a little faster, and sometimes it&#39;s a little slower. \
Mostly within the realm of statistical noise (especially with each query having a \
sample size of only 13, on a VM that lives on a probably-busy \
host).</div><div><br></div><div dir="ltr">On Mon, Jan 29, 2024 at 9:18 AM Ron \
Johnson &lt;<a href="mailto:ronljohnsonjr@gmail.com" target="_blank" \
rel="noreferrer">ronljohnsonjr@gmail.com</a>&gt; wrote:<br></div><div \
class="gmail_quote"><blockquote class="gmail_quote" style="margin:0px 0px 0px \
0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div \
dir="ltr"><div>Yes, jit=on.</div><div><br></div><div>I&#39;ll test them with jit=off, \
to see the difference.   (The application is 3rd party, so will change it at the \
system level.)</div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On \
Mon, Jan 29, 2024 at 7:09 AM Bob Jolliffe &lt;<a \
href="mailto:bobjolliffe@gmail.com" target="_blank" \
rel="noreferrer">bobjolliffe@gmail.com</a>&gt; wrote:<br></div><blockquote \
class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid \
rgb(204,204,204);padding-left:1ex"><div dir="ltr">Out of curiosity, is the pg14 \
running with the default  jit=on setting?    <div><br></div><div>This is obviously \
entirely due to the nature of the particular queries themselves, but we found that \
for our workloads that pg versions greater  than 11 were exacting a huge cost due to \
the jit compiler.   Once we explicitly turned jit=off we started to see improvements. \
</div></div> </blockquote></div></div>
</blockquote></div></div>
</blockquote></div>



[prev in list] [next in list] [prev in thread] [next in thread]