There IS some information on reads versus write in 42-6. I'm hazy on the 
details but it could be useful with the "what is Disc?" question.

Cheers, Martin

Martin Packer,
zChampion, Principal Systems Investigator,
Worldwide Banking Center of Excellence, IBM

+44-7802-245-584

email: [email protected]

Twitter / Facebook IDs: MartinPacker
Blog: 
https://www.ibm.com/developerworks/mydeveloperworks/blogs/MartinPacker



From:   "Mackenzie, Bruce" <[email protected]>
To:     [email protected], 
Date:   04/19/2013 01:27 PM
Subject:        Re: SMF42 subtype 6 interpretation
Sent by:        IBM Mainframe Discussion List <[email protected]>



In the SMF manual above the Data Set I/O Statistics Section for the 
Subtype 6, it does say 

"Note: I/O response and service time components are recorded in multiples 
of 128 micro-seconds."

So XDBDO0 has an average response time of 22.875ms of which 5ms is connect

time and 17ms disconnect time. 

The high connect time would indicate the reading (or possibly the writing)

of multiple pages at a time (index or table space scan).  The disconnect 
time of 17ms could also be considered high.  Disconnect time is normally 
from random reads or if you are synchronously mirroring these devices via 
PPRC it can be from writes.  So there is not a "correct" average response 
time, as it depends on the type of I/O being executed and your 
environment.

Typically the connect time for a single 4KB data page either read or write

should be under 1ms for a modern control unit with a 2Gb FICON channel.  A

non-mirrored write will generally be a cache hit and therefor will not 
have any disconnect time.  A mirrored write will have some disconnect time

based on the size of the block, the speed of the link and the distance 
between the primary and secondary.  A random read will have some 
disconnect time as the control unit pulls the page from the disk, and that

time would depend on how busy the control unit / raid-rank was at the 
time. 

  Thanks



-----Original Message-----
From: IBM Mainframe Discussion List [mailto:[email protected]] On 
Behalf Of Ducky Duc
Sent: Thursday, April 18, 2013 9:54 AM
To: [email protected]
Subject: SMF42 subtype 6 interpretation

Hi all , 

As a DB2 specialist , i have reports showing a wait for io  time in some 
programs. I did a SMF42 subtype 6 reports and , effectively, my datasets 
are in  the top of the list , cut and pasted below from an Excel sheet. 
However ,  don't know if my calculation of the average response time of 9 
; which corresponds to 9 X 128ms = 1152ms , from the manual is correct. 
And sorry for the dummy question , what is a "correct" average response 
time , a few ms i suppose ...

Thank you 
Duc 

DATASET                       VOLSER             RESPTIME IOCOUNT  
AVGCONCT                 AVGPEND                 AVGDISC
DB2XXF0.DSNDBD.F4DB002.F4SS102.I0001.A001                XDBDO0 22.875  
5379               5.187                   0.125                  17.250
DB2XXF0.DSNDBD.F3DB001.F3SS039.I0001.A001                XDBDM5 21.625  
1141              16.750                   0.125                   4.437
DB2XXF0.DSNDBD.F3DB002.F3SS113.I0001.A001                XDBD8A 14.062  
7489               1.875                   0.125                  11.625
DB2XXF0.DSNDBD.F3DB002.IF3TADR1.I0001.A001               XDBI16   12.875

2057               1.750                   0.125                  10.750
DB2XXF0.DSNDBD.F3DB001.F3SS037.I0001.A001                XDBDH3 9.562  
3946               6.125                   0.000                   3.187
DB2XXF0.DSNDBD.F3DB001.IF3T1BGJ.I0001.A001               XDBI08   9.562  
2038               5.625                   0.000                   3.562
DB2XXF0.DSNDBD.F3DB006.F3PS006.I0001.A003                XDBD53 9.187  
1244               5.875                   0.125                   2.937
DB2XXF0.DSNDBD.F3DB006.F3PS006.I0001.A007                XDBDC0 9.062  
1203               5.750                   0.125                   2.937
DB2XXF0.DSNDBD.F3DB006.F3PS006.I0001.A008                XDBDF6 9.062  
1321               6.000                   0.125                   2.687
DB2XXF0.DSNDBD.F3DB006.F3PS006.I0001.A009                XDBDKD 9.062  
1321               5.750                   0.125                   3.062
DB2XXF0.DSNDBD.F3DB006.F3PS006.I0001.A010                XDBDI2   9.062  
1328               6.250                   0.125                   2.375
DB2XXF0.DSNDBD.F3DB001.IF3TCCTD.I0001.A001               XDBI4C   9.062  
1057               5.625                   0.000                   3.187
DB2XXF0.DSNDBD.F5DB001.F5SS012.I0001.A001                XDBDG0 9.062  
1926               7.000                   0.125                   1.625

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions, send email

to [email protected] with the message: INFO IBM-MAIN

NOTICE: Confidential message which may be privileged. Unauthorized 
use/disclosure prohibited. If received in error, please go to 
www.td.com/legal for instructions.
AVIS : Message confidentiel dont le contenu peut être privilégié. 
Utilisation/divulgation interdites sans permission. Si reçu par erreur, 
prière d'aller au www.td.com/francais/avis_juridique pour des 
instructions.

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN







Unless stated otherwise above:
IBM United Kingdom Limited - Registered in England and Wales with number 
741598. 
Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU






----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to