| From: | Claudio Freire <klaussfreire(at)gmail(dot)com> |
|---|---|
| To: | Vitaliy Garnashevich <vgarnashevich(at)gmail(dot)com> |
| Cc: | hzzhangjiazhi <hzzhangjiazhi(at)corp(dot)netease(dot)com>, Gary Doades <gpd(at)gpdnet(dot)co(dot)uk>, Rick Otten <rottenwindfish(at)gmail(dot)com>, "pgsql-performance(at)lists(dot)postgresql(dot)org" <pgsql-performance(at)lists(dot)postgresql(dot)org> |
| Subject: | Re: effective_io_concurrency on EBS/gp2 |
| Date: | 2018-02-05 20:14:53 |
| Message-ID: | CAGTBQpaOsC4=hwtO-ZkE9XVuUS2DqfiLw3-Zf_tjj1CfT49fjw@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-performance |
On Mon, Feb 5, 2018 at 8:26 AM, Vitaliy Garnashevich
<vgarnashevich(at)gmail(dot)com> wrote:
>> I mean, that the issue is indeed affected by the order of rows in the
>> table. Random heap access patterns result in sparse bitmap heap scans,
>> whereas less random heap access patterns result in denser bitmap heap
>> scans. Dense scans have large portions of contiguous fetches, a
>> pattern that is quite adversely affected by the current prefetch
>> mechanism in linux.
>>
>
> Thanks for your input.
>
> How can I test a sparse bitmap scan? Can you think of any SQL commands which
> would generate data and run such scans?
>
> Would a bitmap scan over expression index ((aid%1000)=0) do a sparse bitmap
> scan?
If you have a minimally correlated index (ie: totally random order),
and suppose you have N tuples per page, you need to select less (much
less) than 1/Nth of the table.
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Thomas Güttler | 2018-02-06 14:31:27 | Details after Load Peak was: OT: Performance of VM |
| Previous Message | Justin Pryzby | 2018-02-05 19:19:22 | Re: postgresql 10.1 wrong plan in when using partitions bug |