| From: | Zdenek Kotala <Zdenek(dot)Kotala(at)Sun(dot)COM> |
|---|---|
| To: | Greg Smith <gsmith(at)gregsmith(dot)com> |
| Cc: | Gregory Stark <stark(at)enterprisedb(dot)com>, pgsql-hackers(at)postgresql(dot)org |
| Subject: | Re: Initial prefetch performance testing |
| Date: | 2008-09-23 12:26:58 |
| Message-ID: | 48D8E092.2020301@sun.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Greg Smith napsal(a):
> On Mon, 22 Sep 2008, Gregory Stark wrote:
>
>> I'm quite surprised Solaris doesn't support posix_fadvise -- perhaps
>> it's in some other version of Solaris?
Solaris has only fake variant of posix_fadvise. See
http://src.opensolaris.org/source/xref/onnv/onnv-gate/usr/src/lib/libc/port/gen/posix_fadvise.c
UFS has own optimization. For example if it detects sequential scan then
file cache is limited on 80kB. Or it also has prefetching.
ZFS has intelligent read patter recognitions algorithms and other
improvements.
Zdenek
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Tom Lane | 2008-09-23 12:29:56 | Re: Proposal: move column defaults into pg_attribute along with attacl |
| Previous Message | Bruce Momjian | 2008-09-23 12:23:43 | Re: Proposed patch: make SQL interval-literal syntax work per spec |