Internetworkers: http://www.ibiblio.org/internetworkers/

["Michael Best" <mbest AT dasya.com>]

I usually solve this by using subqueries.  Most databases let you nest
queries.  Doing so allows you to reference calculated columns in the outer
query:

SELECT iteration, label, read_ops, read_data, read_latency, read_ops *
read_data AS concurrency from (SELECT ... ) q1;

You can use the IFNULL function if you want to avoid NULL values:
IFNULL(columnname, 0) will return 0 if columnname is NULL.

-- Michael Best

> -----Original Message-----
> From: internetworkers-bounces AT lists.ibiblio.org 
> [mailto:internetworkers-bounces AT lists.ibiblio.org] On Behalf 
> Of Scott Russell
> Sent: Monday, February 25, 2008 5:23 PM
> To: Internetworkers: http://www.ibiblio.org/internetworkers/
> Subject: Re: [internetworkers] SQL Help?
> 
> 
> On Mon, 2008-02-25 at 18:47 -0500, Scott Russell wrote:
> > On Mon, 2008-02-25 at 16:41 -0500, Ilan Volow wrote:
> > > I'm a PostgreSQL guy, so take my educated guess with a grain of 
> > > salt, but you might want to try this just to see if it works:
> > 
> > Much appreciated. I'll take a look at it...
> > 
> 
> FYI, You're suggestion worked fine. I'm not entirely 
> convinced I'm doing this sanely still, but it does work.
> 
> I also do this to calculate a read_size column which has the 
> formula of (read_bytes/read_ops = read_size). Note that 
> read_ops could be 0 which would yield a NULL in the read_size 
> column. To avoid that, I use a nested IF!
> 
> IF(SUM(IF(latx_stats_data.statname = 'read_ops', `statval`, 
> 0)), (SUM(IF(latx_stats_data.statname = 'read_data', 
> `statval`, 0)) / SUM(IF(latx_stats_data.statname = 
> 'read_ops', `statval`, 0))), 0) AS 'read_size'
> 
> I have to do this with several columns. I wonder if should 
> read up on stored procedures...
> 
> --
> Scott <catfather AT donpoo.net>
> AIM: BlueCame1
> 
> ---
> Come and play at the InterNetWorkers Web site!  
> http://www.ibiblio.org/internetworkers/
> You are currently subscribed to InterNetWorkers mailing list 
> To unsubscribe visit 
> http://lists.ibiblio.org/mailman/listinfo/internetworkers
> 
>