I see no performance impact from your propose increase in JOBNUM.

Be sure your checkpoint dataset is large enough to hold the increased
number of jobs.
/$DCKPTSPACE. If not, define new CKPT's  and use the CHECKPOINT
Migration dialog.

Range (IMO) should be at large enough to hold all possible jobs.
However, I don't think JES requires this (I could be wrong).
What happens if you have more jobs than range, I don't know. In my case
RANGE exceeds JOBNUM.

I recommend nnnnnn = 099999 for RANGE= (up from 36000). If you go beyond
099999, the prefix on the JOBnnnnn in HASP Messages will change to JO or
J and any automation /programs that depend on "JOB" will need to be
recoded.

Of course, it just might be easier to purge a few jobs.....<G>

HTH.

<snip>
*EAX0233I $HASP050 - $HASP050 JES2 RESOURCE SHORTAGE OF JNUM - 91%
UTILIZATION REACHED

  
 $JD DETAILS

$HASP9103 D DETAIL 013                                     
$HASP9104 JES2 RESOURCE USAGE SINCE 2011.040  0:00:00      
RESOURCE    LIMIT    USAGE      LOW     HIGH  AVERAGE      
-------- -------- -------- -------- -------- --------      
BERT       100000     1579     1493     1587     1541      
BSCB          512        0        0        0        0      
BUFX          500        3        1        5        1      
CKVR           17        0        0        1        0      
CMBS         1209        0        0       60        0      
CMDS         1200        0        0        0        0      
ICES          387        0        0        0        0      
JNUM        37767    34369    34316    34369    34350      
JOES        50000    43599    43551    43599    43582      
JQES        36500    34369    34316    34369    34350      
LBUF          500        0        0        0        0      
NHBS          204        0        0        0        0      
SMFB           48        0        0        0        0      
TBUF          104        0        0        0        0      
TGS        183464   134701   134583   134701   134651      
TTAB            3        0        0        0        0      
VTMB          512        0        0        0        0      
$HASP9105 JES2 SAMPLING STATISTICS SINCE 2011.040  0:00:00 
TYPE               COUNT  PERCENT                          
----------------  ------  -------                          
ACTIVE                18     0.32                          
WAITING DISPATCH       1     0.01                          
IDLE                5602    99.62                          
LOCAL LOCK             0     0.00                          
NON-DISPATCHABLE       0     0.00                          
PAGING                 2     0.03                          
OTHER WAITS            0     0.00                          
TOTAL SAMPLES       5623                                   

In our JES2 parm we have the following value;
JOBDEF         ACCTFLD=IGNORE,   
               JCLERR=YES,       
               JOBNUM=36500,     
               JNUMWARN=90,      
               JOBWARN=80,       
               PRTYHIGH=9,       
               PRTYJECL=NO,      
               PRTYJOB=YES,      
               PRTYLOW=5,        
               PRTYRATE=0,       
               RANGE=1-37767     

 Does anyone know of any system degradation if we up the value of the
JOBNUM=36500 by 20% or so, is there anything we should consider before
changing the value?  Do we need to consider increasing the RANGE value
as well as the JOBNUM value? 
</snip>

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

Reply via email to