It is my cold and fuzzy recollection that 6144 was in vogue for a few years 
when that value was deemed to be VERY good on a 3340 DASD and not too shabby 
[1] for most other flavors. 
Bill Fairchild 
  
[1] Or, more precisely, "okay" or "swell". 

----- Original Message -----

From: "Elardus Engelbrecht" <[email protected]> 
To: [email protected] 
Sent: Wednesday, August 13, 2014 1:57:33 PM 
Subject: Re: Extents more than One for load modules library 

John Eells wrote: 

> I hope the formatting will survive (we'll see). 

It did! 

[ lots of interesting things snipped ] 

> Unless you're really into pain, I suggest *not* asking her to give you the 
> numbers.  They'll give you a headache. Really. 

For once I will listen to you!!! 

[ more snippage ] 

>  These contain the executable code, funny-looking machine language stuff. 

Hey! My assembler programs are not funny even when they're bug free... 

> I tested one data set in 1999, for DFSORT, and found a 20% reduction in space 
> utilization when the library was blocked at 32760 vs. 6144.  20% is 
> significant. 

Interesting. Thanks. Hmmm, 6144, wasn't that used a lot for SAS work files? 

>This takes 14ms on native 3390 DASD.  This is 1,673 Dog Years to a computer. 

I'm doggone mad! Why or how did you came to that dog years? 

[ more snippage! My scissors are not that sharp anymore! ;-D ] 

> So larger block sizes still mean better performance. 

That is a relief! Here I agree with you! 

John, Many thanks for your very good and educational post! I really appreciate 
it very much! 

Groete / Greetings 
Elardus Engelbrecht 

---------------------------------------------------------------------- 
For IBM-MAIN subscribe / signoff / archive access instructions, 
send email to [email protected] with the message: INFO IBM-MAIN 


----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to