256 was not the limit in 1994 when I tried to run some code that ATTACHed a new 
subtask for each DASD device in a system that had thousands and thousands of 
DASDs.  Shortly after my test job started, I realized my grand design error and 
canceled the  job.  It took a very large number of minutes for the job to 
terminate all those subtasks abnormally, process them with RTM, dump their 
control blocks, etc. 

Bill Fairchild 
Franklin, TN 

“Political language is designed to make lies sound truthful and murder 
acceptable, and to give the appearance of solidity to pure wind.” [George 
Orwell] 

----- Original Message -----
From: "Shmuel Metz (Seymour J.)" <[email protected]> 
To: [email protected] 
Sent: Friday, April 19, 2013 10:10:44 AM 
Subject: Re: Idle TCBs -- performance question 

In 
<b870629719727b4ba82a6c06a31c291239b4fc3...@hqmailsvr01.voltage.com>, 
on 04/18/2013 
   at 10:02 AM, Phil Smith <[email protected]> said: 

>So the question is: What's the real cost of having a TCB sitting 
>there idle? 

Slightly more LSQA. Slightly longer queue length. 

>To what extent is an idle TCB on z/OS impacting things? 

A longer TCB queue slows dispatching, but not by that much. 

>Are there system- and/or primary TCB-based limits on numbers of 
>TCBs that we're pushing against? 

How close are you to 256? Is that still the limit? 

>Each TCB uses maybe 25K of memory in our address space, a 

What kind of memory? 

>and presumably some small amount of system memory. 

LSQA is in your address space. Im not aware of any per TCB global 
memory use. 

-- 
     Shmuel (Seymour J.) Metz, SysProg and JOAT 
     Atid/2        <http://patriot.net/~shmuel> 
We don't care. We don't have to care, we're Congress. 
(S877: The Shut up and Eat Your spam act of 2003) 

---------------------------------------------------------------------- 
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