High faulting in the System Memory Pool

["Graap, Ken" <keg@xxxxxxxxxxxxx>]

I'm having an event occur on my system this morning that I'm still trying to
understand.

The faulting rate in the system *MACHINE memory pool skyrocketed.

I've seen this happen before when a communications recovery is in progress,
but I checked that out and our comm interfaces seemed to be performing just
fine.

I did notice that a couple batch jobs were doing big query sorts (QQQTSORT)
.. and when they ended the faulting rate tapered off for a few minutes. But
it is still at about 9.8/second with an allocation of 11GB... That is high.
We usually can run at 2 faults/second with about 2GB allocated. 

Does anyone have any ideas on what I might look at to determine what is
going on here?

     Pool   Reserved    Max  ----DB-----  --Non-DB---  Act-  
    Size M   Size M     Act  Fault Pages  Fault Pages  Wait  
  12368.42    672.18  +++++     .3    .3    9.4  10.9  281.0 
   2765.47      1.65    180   17.7 129.1   54.8 306.4   3855 
    655.35     12.01     15   86.9  7946   49.3 928.2    4.6 
    423.93       .00     25   92.5 729.7  159.3 492.0  586.7 
      6.96       .00      6     .0    .6    7.6  26.3   15.1 
     81.91       .00      4    1.6 106.5   13.5  27.6    9.8 
     81.91       .00      8   10.5  96.5   26.6  94.6   67.3 

thanx

Kenneth
--
********************************
Kenneth  E.  Graap
IBM Certified Specialist
AS/400 Professional
System Administrator
NW Natural - Information Services
System Services
Phone 503 226 4211 X5537
FAX   603 849 0591
keg@xxxxxxxxxxxxx
********************************