Execution plan, I/O, CPU cost?

Hey there ...
I'm trying to figure out what these numbers can tell me.
Running this one -
select proptrans.proptranstext as groupname
from propval
inner join proptrans
on propval.propvalguid = proptrans.propvalguid
where propval.propownerrefguid = '12121'
and propval.propguid = 'name'
Gives me 6 in I/O, 7 in CPU cost, in a clustered index seek, on both
propval, and proptrans.
What does these numbers tell me? Is it slow, fast etc?
I doubt, therefore I might be.
What is propval and proptrans?
Simon Worth
Kim Noer wrote:
> Hey there ...
> I'm trying to figure out what these numbers can tell me.
> Running this one -
> select proptrans.proptranstext as groupname
> from propval
> inner join proptrans
> on propval.propvalguid = proptrans.propvalguid
> where propval.propownerrefguid = '12121'
> and propval.propguid = 'name'
> Gives me 6 in I/O, 7 in CPU cost, in a clustered index seek, on both
> propval, and proptrans.
> What does these numbers tell me? Is it slow, fast etc?
>
|||Kim Noer wrote:
> Hey there ...
> I'm trying to figure out what these numbers can tell me.
> Running this one -
> select proptrans.proptranstext as groupname
> from propval
> inner join proptrans
> on propval.propvalguid = proptrans.propvalguid
> where propval.propownerrefguid = '12121'
> and propval.propguid = 'name'
> Gives me 6 in I/O, 7 in CPU cost, in a clustered index seek, on both
> propval, and proptrans.
> What does these numbers tell me? Is it slow, fast etc?
The important thing is that you are seeing a clustered index seek, which
is good. I wouldn't worry about the numbers themselves. The cost only
helps, in my opinion, when you run more than one version of a query in
the same batch. Query Analyzer shows you are cost differential between
the queries.
David Gugick
Imceda Software
www.imceda.com
|||"David Gugick" <davidg-nospam@.imceda.com> wrote in message
news:%233P9mGELFHA.1284@.TK2MSFTNGP14.phx.gbl

> The important thing is that you are seeing a clustered index seek,
> which is good. I wouldn't worry about the numbers themselves. The
> cost only helps, in my opinion, when you run more than one version of
> a query in the same batch. Query Analyzer shows you are cost
> differential between the queries.
I don't worry about the cost numbers, but it's slightly annoying to me, that
they are there, and there isn't really any explaination on them at all.
More and one version, with that you mean the operation being done, but in
different ways (like joins vs. select in etc etc)?
I doubt, therefore I might be.