Re: A question about Vacuum analyze - Mailing list pgsql-general

From Emi Lu
Subject Re: A question about Vacuum analyze
Date
Msg-id 44033B71.5030204@encs.concordia.ca
Whole thread Raw
In response to Re: A question about Vacuum analyze  (Christopher Browne <cbbrowne@acm.org>)
List pgsql-general
Thank you very much for all your inputs. I believe "analyze" is the one
I should use .

>Quoth emilu@encs.concordia.ca (Emi Lu):
>
>
>>>no. the suggestion was that a VACUUM is not needed, but that an
>>>ANALYZE might be.
>>>
>>>
>>Thank you gnari for your answer. But I am a bit confused about not
>>running vacuum but only "analyze". Can I seperate these two
>>operations? I guess "vacuum analyze" do both vacuum and analyze. Or
>>"EXPLAIN ANALYZE" can do it for me?
>>
>>
>
>EXPLAIN, ANALYZE, and VACUUM are different things; ANALYZE gets used
>in two different contexts...
>
>1.  VACUUM is what cleans dead tuples out of tables.
>
>  e.g. VACUUM my_table;
>
>2.  VACUUM ANALYZE cleans out dead tuples and recalculates data
>    distributions
>
>  e.g. VACUUM ANALYZE my_table;
>
>3.  EXPLAIN describes query plans
>
>  e.g. EXPLAIN select * from my_table;
>
>4.  EXPLAIN ANALYZE compares query plan estimates to real results
>
>  e.g. EXPLAIN ANALYZE select * from my_table;
>
>5.  ANALYZE recalculates data distributions (as in 2, but without
>    cleaning out dead tuples).
>
>  e.g. ANALYZE my_table;
>
>Pointedly, EXPLAIN ANALYZE is entirely distinct from ANALYZE and
>VACUUM ANALYZE...
>
>


pgsql-general by date:

Previous
From: Bernhard Weisshuhn
Date:
Subject: Re: ltree + gist index performance degrades significantly over a night
Next
From: CG
Date:
Subject: Re: ltree + gist index performance degrades significantly over a night