Re: Bytea poor performance

From: "NSO" <nso(at)fmf(dot)vtu(dot)lt>
To: "Sean Davis" <sdavis2(at)mail(dot)nih(dot)gov>
Cc: pgsql-performance(at)postgresql(dot)org
Subject: Re: Bytea poor performance
Date: 2005-10-15 14:48:54
Message-ID: 17532.81.7.89.65.1129387734.squirrel@fmf.vtu.lt
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance


Yes, it takes even up to 35 seconds.. I did the same query on the server
(not PC with was connected directly to server with 100mbit net), and /I
got better result it is 3.5 - 4 seconds, but it still not good.. Why it
is slow? and why the difference is so big? I mean from 4 to 35 seconds?

thx

> On 10/15/05 10:00 AM, "NSO" <nso(at)fmf(dot)vtu(dot)lt> wrote:
>
>> Hello,
>>
>> How about some explain analyze output?
>> Explain analyse select * from files where nr > 1450
>>
>> "Index Scan using pk on files (cost=0.00..3.67 rows=50 width=36)
>> (actual time=0.000..0.000 rows=50 loops=1)"
>
> I may not be understanding the output, but your actual time reports 0 for
> the query. And the total runtime is 23 seconds?
>
> Sean
>
>
>>> On 10/15/05 9:20 AM, "NSO" <nso(at)fmf(dot)vtu(dot)lt> wrote:
>>>
>>>> Hello,
>>>>
>>>> I am trying to select form table with bytea field. And queries runs
>>>> very
>>>> slow.
>>>> My table:
>>>> CREATE TABLE files (file bytea, nr serial NOT NULL) WITH OIDS;
>>>>
>>>> Query:
>>>> select * from files where nr > 1450
>>>>
>>>> (I have total 1500 records in it, every holds picture of 23kB size)
>>>> Query runs very long:
>>>> Total query runtime: 23625 ms.
>>>> Data retrieval runtime: 266 ms.
>>>> 50 rows retrieved.
>>>>
>>>> explain:
>>>> Index Scan using pk on files (cost=0.00..3.67 rows=50 width=36)
>>>> Index Cond: (nr > 1450)
>>>>
>>>> Is it possible to do something with it? or it is normal? Our server is
>>>> fast, and all other tables work fine..
>>>
>>> How about some explain analyze output? Have you done a full vacuum
>>> lately?
>>> How about reindexing?
>>>
>>> Sean
>>>
>>>
>>> ---------------------------(end of
>>> broadcast)---------------------------
>>> TIP 6: explain analyze is your friend
>>>
>>> --
>>> This message has been scanned for viruses and
>>> dangerous content, and is believed to be clean.
>>>
>>>
>>
>>
>
>
> --
> This message has been scanned for viruses and
> dangerous content, and is believed to be clean.
>
>

--
This message has been scanned for viruses and
dangerous content, and is believed to be clean.

In response to

Browse pgsql-performance by date

  From Date Subject
Next Message Tom Lane 2005-10-15 15:54:07 Re: Bytea poor performance
Previous Message Craig A. James 2005-10-15 14:47:08 tsearch2/GIST performance factors?