[thesite] blockfactor revisted

[Daniel J. Cody]

rudy, how would we run a query to see how much data its bringing back?

i'd be more interested to know about the specifics that Allaire or 
whoever bases these recomendations on..

e.g. is it better to go with 32K thresholds if you have 1000000 rows and 
a 4 way xeon box?

i've tried prying these kind of details out of people like ray btw. IP, 
trade secrets, whatever. <insert rant about closed software here> .. 
again. :)

.djc.

.jeff wrote:

> dan,
> 
> 
>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
>>From: Daniel J. Cody
>>
>>a range between 70 - 100 or 35-45 looks to be best.
>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
>>
> 
> i'm actually more curious about the average number of bytes per record on
> the getarticles query.  the more i read about blockfactor, the more i'm
> convinced we need to watch that 32K threshold.