HP3000-L Archives

June 1997, Week 1

HP3000-L@RAVEN.UTC.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Stan Sieler <[log in to unmask]>
Reply To:
Stan Sieler <[log in to unmask]>
Date:
Thu, 5 Jun 1997 13:32:33 -0700
Content-Type:
text/plain
Parts/Attachments:
text/plain (40 lines)
David asks:
> unable to find any threads in previous posts.  So, I'll ask...How much
> overhead is there when running Image logging?  CPU, disk, memory,
> response times, etc.

I haven't measured disk/memory/response time, but I *did* measure CPU
overhead (not counting overhead incurred by interrupt processing),
and saw about 33% increased CPU usage.  YMMV, of course.
(Measurement on MPE/iX 5.0)

DBPUTs to a small detail data set, linked to a single automatic
master), showing count of number of PA-RISC instructions:

without logging:                   with logging:

               #Instructions                      #Instructions
   KEY1 =>> 1   94856                KEY1 =>> 2       ?
   KEY1 =>> 2   27490                KEY1 =>> 2       ?

   KEY1 =>> 3   25482                KEY1 =>> 3   39325
   KEY1 =>> 4   30293                KEY1 =>> 4   43217
   KEY1 =>> 5   28837                KEY1 =>> 5   42718
   KEY1 =>> 6   29975                KEY1 =>> 6   41916
   KEY1 =>> 7   30978                KEY1 =>> 7   46193
   KEY1 =>> 8   32545                KEY1 =>> 8   44951
   KEY1 =>> 9   30530                KEY1 =>> 9   42214
   KEY1 =>>10   92691                KEY1 =>>10  105428
   KEY1 =>>11   10039 (full)         KEY1 =>>11   10426 (full)
   ------------------                ------------------
   Sum 3..11:  311370                Sum 3..11:  416388

Ratio:  1.33      (Cost of dbputs 3..11  with/without)

However, that seems like it might be a low price for the benefits of
logging.

--
Stan Sieler                                          [log in to unmask]
                                     http://www.allegro.com/sieler.html

ATOM RSS1 RSS2