| Title: | Zap Technical Conference |
| Notice: | ZAP Version 5.3 is available. See note 1.1 |
| Moderator: | ZAPDEV::MACONI |
| Created: | Mon Feb 24 1986 |
| Last Modified: | Mon May 05 1997 |
| Last Successful Update: | Fri Jun 06 1997 |
| Number of topics: | 170 |
| Total number of notes: | 492 |
System Overhead of ZAP
My customer would like to know what type of system overhead he
can expect with the implementation of ZAP. He has a 8550 with a
terminal network of 100 devices. He will use ZAP only as a way
of logging off users after a specific inactive time period.
Thanks for any assistance you might be able to give.
Karen Guenther
| T.R | Title | User | Personal Name | Date | Lines |
|---|---|---|---|---|---|
| 55.1 | ZAP V3.8 Performance | MPO::MACONI | Thu Dec 22 1988 14:05 | 27 | |
Overhead for ZAP is extremely low. Below are some numbers that
I have taken from the systems at our site for comparison:
CPU Type # passes(+) # stopped Ave # Proc CPU time Page Flts
8350 1,294 0 35 4:28 204
8350 25,664 93 40 2:00:16 219
8550 17,411 507 90 25:27 251
8700 17,399 498 90 28:20 250
8700 1,791 209 65 2:36 233
Physical Memory usage averages 375 pages.
(+) There is one pass per minute made.
From these numbers, you can see that the average amount of CPU time
used in each 60 passes (1 hour) is approximately 5.3 seconds per
hour on the larger 8xxx systems, and 12 to 16 seconds per hour on
the smaller systems. Out of 3600 seconds per hour, this means overhead
of 0.14% and 0.45% respectively.
Please note that these numbers where just gathered on the fly and
were not part of an actual performance evaluation. Performance
is more a function of the number of rules in the database than of
the number of processes on the system.
Keith Maconi
| |||||