Sample output from the MVS Component |
|---|
Please note that the output is narrative. The software analyzes data in your performance data base, and produces a narrative report much as would a performance analyst.
The MVS Component User Manual could be consulted for a detailed description
of each finding produced by the software. A sample "rule" description
from the MVS Component User Manual can be reviewed to appreciate the
level of documentation provided with the MVS Component.
MVS PERFORMANCE CONSTRAINTS ON 02OCT1996 - SYSTEM: A431
Analyzed by Version 7.1.0 of CPExpert(tm)
(C) Copyright 1997, Computer Management Sciences, Inc.
PAGE: 12
RULE MVS602: XCF MESSAGE BUFFER LENGTH MAY BE TOO SMALL
DEFAULT: The XCF message buffers may be too small for this transport class
Unnecessary overhead is incurred when XCF must expand the buffers to fit
a message. The CLASSLEN parameter was specified as 956 for this
transport class. You should consider increasing the message length
for this transport class or you may wish to split the transport class,
depending upon actual message lengths. This finding applies to the
following RMF measurement intervals:
SENT SMALL MESSAGES MESSAGES TOTAL
MEASUREMENT INTERVAL TO MESSAGES THAT FIT TOO BIG MESSAGES
8:00- 8:15,02OCT1996 CPU2 0 2,393 194 2,587
8:00- 8:15,02OCT1996 CPU4 0 65,810 814 66,624
8:15- 8:30,02OCT1996 CPU2 0 2,281 174 2,455
8:30- 8:45,02OCT1996 CPU2 0 2,165 135 2,300
8:45- 9:00,02OCT1996 CPU2 0 2,076 104 2,180
9:00- 9:15,02OCT1996 CPU2 0 2,342 112 2,454
9:15- 9:30,02OCT1996 CPU2 0 2,554 132 2,686
RULE MVS605: THE XCF INBOUND MESSAGE BUFFER SPACE MAY BE TOO SMALL
The inbound message buffer space may be too small. CPExpert noticed
that XCF input requests were rejected because of constraints on the
amount of input message buffer space. An asterisk beside the buffer
space means that the buffer space DECREASED during the reported
measurement interval, from the preceeding measurement interval. You
should consider increasing the amount of input message buffer space.
This finding applies to the following measurement intervals:
RECEIVED TOTAL REJECTED PCT BUFFER
MEASUREMENT INTERVAL FROM REQUESTS REQUESTS REJECTED SPACE
11:30-11:45,02OCT1996 CPU4 4 4 100.0 2,048K
RULE MVS651: LOCK CONTENTION WAS HIGH
DB2DBP1_LOCK1: The lock contention for this structure was higher than
normal. Higher lock contention can result in an increase in central
processor utilization and a reduction in throughput. If this finding
continues to occur, you should review the alternatives listed in the
WLM Component User Manual. If you are unable to take action, you
should consider increasing the LOCKCONT guidance variable, located
in USOURCE(MVSGUIDE). The LOCKCONT variable currently is 2%.
TOTAL LOCK REQUESTS WITH PERCENT LOCK
MEASUREMENT INTERVAL REQUESTS LOCK CONTENTION CONTENTION
8:00- 8:15,02OCT1996 9,740 554 6
8:15- 8:30,02OCT1996 1,103 32 3
8:30- 8:45,02OCT1996 693 16 2
9:30- 9:45,02OCT1996 242 11 5
9:45-10:00,02OCT1996 442 41 9
10:00-10:15,02OCT1996 282 12 4
10:15-10:30,02OCT1996 319 17 5
12:15-12:30,02OCT1996 454 22 5
MVS PERFORMANCE CONSTRAINTS ON 02OCT1996 - SYSTEM: A431
Analyzed by Version 7.1.0 of CPExpert(tm)
(C) Copyright 1997, Computer Management Sciences, Inc.
PAGE: 13
RULE MVS651: LOCK CONTENTION WAS HIGH
JES2_CKPT1: The lock contention for this structure was higher than
normal. Higher lock contention can result in an increase in central
processor utilization and a reduction in throughput. If this finding
continues to occur, you should review the alternatives listed in the
WLM Component User Manual. If you are unable to take action, you
should consider increasing the LOCKCONT guidance variable, located
in USOURCE(MVSGUIDE). The LOCKCONT variable currently is 2%.
TOTAL LOCK REQUESTS WITH PERCENT LOCK
MEASUREMENT INTERVAL REQUESTS LOCK CONTENTION CONTENTION
11:30-11:45,02OCT1996 15,183 313 2
14:30-14:45,02OCT1996 15,036 334 2
15:15-15:30,02OCT1996 13,934 286 2
15:30-15:45,02OCT1996 17,326 488 3
16:30-16:45,02OCT1996 14,910 304 2
17:00-17:15,02OCT1996 26,769 690 3
17:15-17:30,02OCT1996 10,397 318 3
RULE MVS652: FALSE LOCK CONTENTION WAS HIGH
DB2DBP2_LOCK1: The number of locks with false contention should be less
than 0.1% of the total requests, based on IBM's guidance. The false
lock contention exceeded the guideline for this structure. False lock
contention occurs when the hashing algorithm hashes to the same lock
table entry (hash value) for two different locks. False lock contention
can be reduced by increasing the size of the lock structure or selecting
a better value for the MAXUSRS parameter in IRLMPROC. Refer to Rule
MVS652 in the MVS Component User Manual for additional suggestions.
TOTAL LOCK FALSE LOCK PERCENT FALSE
MEASUREMENT INTERVAL REQUESTS CONTENTION LOCK CONTENTION
15:15-15:30,02OCT1996 12,676 2,654 21
RULE MVS660: SERVICE TIME WAS HIGH FOR SYNCHRONOUS REQUESTS
DB2DBP1_GBP2: The service time for this structure has exceeded the
guidelines for synchronous requests. Service time is accumulated from
the time MVS issues a command for the coupling facility until the return
from the command is recognized by MVS. Service time is recorded for
each structure used by each system. You can alter the times used by
CPExpert in making this finding by altering the SYNCSRV guidance
variables in USOURCE(MVSGUIDE) if you are unable to make changes to
reduce service time for the structure.
TOTAL SYNC AVERAGE SERVICE
MEASUREMENT INTERVAL REQUESTS TIME (MICROSECONDS)
8:30- 8:45,02OCT1996 13 8,030
9:15- 9:30,02OCT1996 7 7,714
10:15-10:30,02OCT1996 22 7,972
16:00-16:15,02OCT1996 6 8,052
MVS PERFORMANCE CONSTRAINTS ON 02OCT1996 - SYSTEM: A431
Analyzed by Version 7.1.0 of CPExpert(tm)
(C) Copyright 1997, Computer Management Sciences, Inc.
PAGE: 14
RULE MVS661: SERVICE TIME WAS HIGH FOR ASYNCHRONOUS REQUESTS
DB2DBP1_GBP0: The service time for this structure has exceeded the
guidelines for asynchronous requests. Service time is accumulated from
the time MVS issues a command for the coupling facility until the return
from the command is recognized by MVS. Service time is recorded for
each structure used by each system. You can alter the times used by
CPExpert in making this finding by altering the ASYNCSRV guidance
variables in USOURCE(MVSGUIDE) if you are unable to make changes to
reduce service time for the structure.
TOTAL ASYNC AVERAGE SERVICE
MEASUREMENT INTERVAL REQUESTS TIME (MILLISECONDS)
12:00-12:15,02OCT1996 1 10.82
RULE MVS661: SERVICE TIME WAS HIGH FOR ASYNCHRONOUS REQUESTS
DB2DBP1_GBP3: The service time for this structure has exceeded the
guidelines for asynchronous requests. Service time is accumulated from
the time MVS issues a command for the coupling facility until the return
from the command is recognized by MVS. Service time is recorded for
each structure used by each system. You can alter the times used by
CPExpert in making this finding by altering the ASYNCSRV guidance
variables in USOURCE(MVSGUIDE) if you are unable to make changes to
reduce service time for the structure.
TOTAL ASYNC AVERAGE SERVICE
MEASUREMENT INTERVAL REQUESTS TIME (MILLISECONDS)
12:00-12:15,02OCT1996 1 10.55
RULE MVS665: TOO MANY SYNCHRONOUS REQUESTS WERE CHANGED TO ASYNCHRONOUS
DB2DBP2_GBP3: The structure experienced too many requests being changed
from synchronous to asnychronous. If MVS determines that a synchronous
request will be significantly delayed (perhaps because subchannels are
busy), MVS will change the request to an asynchronous request (note that
synchronous lock requests are not changed). This finding could indicate
that you need additional coupling facility links.
TOTAL SYNC SYNCH REQUESTS PERCENT
MEASUREMENT INTERVAL REQUESTS CHANGED TO ASYNCH CHANGED
7:15- 7:30,03OCT1996 242 29 12.0