From: | John Naylor <johncnaylorls(at)gmail(dot)com> |
---|---|
To: | Ants Aasma <ants(dot)aasma(at)cybertec(dot)at> |
Cc: | Jeff Davis <pgsql(at)j-davis(dot)com>, Heikki Linnakangas <hlinnaka(at)iki(dot)fi>, Junwang Zhao <zhjwpku(at)gmail(dot)com>, jian he <jian(dot)universality(at)gmail(dot)com>, Andres Freund <andres(at)anarazel(dot)de>, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>, Gurjeet Singh <gurjeet(at)singh(dot)im>, pgsql-hackers(at)postgresql(dot)org |
Subject: | Re: Change GUC hashtable to use simplehash? |
Date: | 2024-02-02 09:21:01 |
Message-ID: | CANWCAZbPOtvqqR7wNFefRD+cwOVDhUFbhq70A6YDGi+hGV6RXg@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On Tue, Jan 30, 2024 at 7:51 PM Ants Aasma <ants(dot)aasma(at)cybertec(dot)at> wrote:
>
> It didn't calculate the same result because the if (mask) condition
> was incorrect. Changed it to if (chunk & 0xFF) and removed the right
> shift from the mask.
Yes, you're quite right.
> It seems to be half a nanosecond faster, but as I
> don't have a machine set up for microbenchmarking it's quite close to
> measurement noise.
With my "throughput-ush" test, they look good:
pgbench -n -T 20 -f bench_cstr_aligned.sql -M prepared | grep latency
master:
latency average = 490.722 ms
(Ants Aantsma) v-17 0001:
latency average = 385.263 ms
v17 0001+0002:
latency average = 339.506 ms
> I didn't post the harness as it's currently so messy to be near
> useless to others. But if you'd like to play around, I can tidy it up
> a bit and post it.
I'd be curious, thanks.
Attachment | Content-Type | Size |
---|---|---|
v17-0002-Shorten-dependency-chain-for-computing-hash-mask.patch | text/x-patch | 2.3 KB |
v17-0001-Speed-up-last-iteration-of-aligned-fasthash.patch | text/x-patch | 2.2 KB |
From | Date | Subject | |
---|---|---|---|
Next Message | Hayato Kuroda (Fujitsu) | 2024-02-02 09:41:35 | RE: speed up a logical replica setup |
Previous Message | Richard Guo | 2024-02-02 09:18:59 | Re: Check lateral references within PHVs for memoize cache keys |