[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference clusta::acms

Title:	ACMS comments and questions
Notice:	This is not an official software support channel. Kits 5.*
Moderator:	CLUSTA::HALLAN

Created:	Mon Feb 17 1986
Last Modified:	Fri Jun 06 1997
Last Successful Update:	Fri Jun 06 1997
Number of topics:	4179
Total number of notes:	15091

4093.0. "Performance of ACMS 4.1 vs V3.2" by VAXSPO::MOREY (Life, adventure without survivors) Tue Jan 28 1997 10:38

	Hi,

	We have a customer now with this configuration:

	 256Mbytes      256Mbytes      256Mbytes      256Mbytes
	----------     ----------     ----------     ----------
	|	 |     |	|     |	       |     |	      |
	| VAX4500|     |VAX4500A|     |VAX4500A|     |VAX4700A|
	|	 |     |	|     |	       |     |	      |
	|	 |     |	|     |	       |     |	      |
	|	 |     |	|     |	       |     |	      |
	----------     ----------     ----------     ----------
	    |		    |		   |		  |
	    |		    |		   |	          |
	------------------------------------------------------
				  |
	  ^                 ^	DISK's    ^              ^
	  |		    |             |              |
	  -------------------		  ----------------
              OpenVMS 6.2		   OpenVMS 5.5-2H4
	      ACMS    4.1	           ACMS    3.2
            ---------------
		 NEW

	Recently the customer upgraded the OpenVMS and ACMS from 5.5-2H4 and 
3.2 to 6.2 and 4.1 respectively of two systems and now he's blaming the 
performance degrate with these new versions. All systems use the same 
application and the same parameters. I'd like to know if the ACMS 4.1 could 
have a worse performance than the 3.2???

	Is there any tool, or method of research to find out where could be the 
problem of these new versions???

	Thanks in advance,


					Carlos Morey

T.R Title User Personal
Name Date Lines

4093.1 OHMARY::HALL Bill Hall - ACMS Engineering - ZKO2-2 Tue Jan 28 1997 12:48 17

T.R	Title	User	Personal Name	Date	Lines
4093.1		OHMARY::HALL	Bill Hall - ACMS Engineering - ZKO2-2	`Tue Jan 28 1997 12:48`	17
	What you are probably seeing is the result of the CMA timers in all of the ACMS processes and their effect on the system. There's a CMA patch that will disable the timers in the processes that don't use CMA (like ACC, EXC and the SPs). The only real place CMA is used is in CP. We are addressing this problem in ACMS V4.2 by removing CMA from all the images that don't explicitly require it. There's some discussion in this notesfile about the timers and even more information in the DECthreads notes conference. Bill

    
    	What you are probably seeing is the result of the CMA timers
    	in all of the ACMS processes and their effect on the
    	system.
    
    	There's a CMA patch that will disable the timers in the processes
    	that don't use CMA (like ACC, EXC and the SPs).  The only real
    	place CMA is used is in CP.
    
    	We are addressing this problem in ACMS V4.2 by removing CMA from
    	all the images that don't explicitly require it.
    
    	There's some discussion in this notesfile about the timers and
    	even more information in the DECthreads notes conference.
    
    	Bill