RDBPROD How to interpret properly the page faults generated by rdbserver process
edited Jan 10, 2014 5:32AM in Rdb on OpenVMS, IBM zSeries and Oracle on OpenVMS (MOSC) 4 commentsAnswered
Hello everyone,
This is a sample of the "show system" command :
OpenVMS V7.3-2 on node AXPDB 18-MAR-2009 11:29:13.46 Uptime 82 16:38:14
Pid Process Name State Pri I/O CPU Page flts Pages
............
00010CFB RDBSERVER_495 LEF 6 13052 0 00:00:01.12 5882 1093 N
00012EFC RDBSERVER_16770 LEF 6 4214 0 00:00:00.47 3779 1107 N
0000ED02 RDBSERVER_9196 LEF 6 574299 0 00:00:36.51 40608 2009 N
00010D04 RDBSERVER_25118 LEF 6 22098 0 00:00:02.04 9699 1450 N
.............
RDBSERVER_9196 process consumed only about37 seconds of processor time and generated 40608. It seems to be too high to me.
So I wanted to compare it with the PFRATH system parameter.
I divide 40608 (page faults generated by the process) by 37 and muliply by 10 (PFRATH is defined in pagefauls per 10 seconds).
The value I get is about 10975.
This is a sample of the "show system" command :
OpenVMS V7.3-2 on node AXPDB 18-MAR-2009 11:29:13.46 Uptime 82 16:38:14
Pid Process Name State Pri I/O CPU Page flts Pages
............
00010CFB RDBSERVER_495 LEF 6 13052 0 00:00:01.12 5882 1093 N
00012EFC RDBSERVER_16770 LEF 6 4214 0 00:00:00.47 3779 1107 N
0000ED02 RDBSERVER_9196 LEF 6 574299 0 00:00:36.51 40608 2009 N
00010D04 RDBSERVER_25118 LEF 6 22098 0 00:00:02.04 9699 1450 N
.............
RDBSERVER_9196 process consumed only about37 seconds of processor time and generated 40608. It seems to be too high to me.
So I wanted to compare it with the PFRATH system parameter.
I divide 40608 (page faults generated by the process) by 37 and muliply by 10 (PFRATH is defined in pagefauls per 10 seconds).
The value I get is about 10975.
0