| Introduction |
This report consists of five parts: - This Introduction and a page of general run statistics. - A listing of the service policy which CPExpert extracted from the SMF Type 72 records. If the service policy was changed, CPExpert will list the new service policy and show the date/time the new policy was activated. - Service Policy Findings (if any). The Service Policy Findings are rules in the WLM001 to WLM050 range. These findings help identify problems or potential problems with the Workload Manager service definition. It is important to realize that these findings normally identify a POTENTIAL problem. Your systems programming staff must decide whether the findings (and their associated recommendations) make sense in your environment. For example, your systems programming staff might have deliberately selected certain parameter values. The values might be appropriate for your installation and your management objectives, even though CPExpert might produce a rule indicating that there is a potential problem with the parameter. You can disable CPExpert's checking the service definition by modifying the CHKPLCY guidance variable in USOURCE(WLMGUIDE). If the CHKPLCY guidance variable is set to N, CPExpert will not check the service definition for potential problems. - General System Findings (if any). The General System Findings are rules in the WLM050 to WLM099 range. These findings identify problems or potential problems with your overall system. For example, many of the rules deal with problems with the paging subsystem. These findings are made only if CPExpert detected that a performance goal was not met and that some general system problem might have caused the goal to be missed. - Specific Findings. The Specific Findings are rules above WLM100. These findings are made if CPExpert detected that a service class did not meet its performance goal. In the Specific Findings, CPExpert attempts to isolate the reason(s) the performance goal was not met. The Specific Findings also include findings related to Cross System Coupling Facility (XCF) performance. The XCF Findings are rules in the WLM600 to WLM699 range. These findings are made when CPExpert detects problems or potential problems with XCF performance. When findings are produced by the WLM Component, please refer to the detailed description of the rule contained in the WLM Component User Manual. The User Manual will describe why the finding was made, give an assessment of the overall performance impact of the finding, and discuss alternatives which may be implemented to solve the problem. Thank you for using CPExpert! We hope that it helps you analyze your system performance under IBM's Workload Manager. Please call Computer Management Sciences at (703) 922-7027 if you have any questions or if you have suggestions for product improvement. |
| Guidance variables |
WORKLOAD MANAGER ANALYSIS, Tue, Mar 20, 2001 - SYSTEM: SYS1
Analyzed by Version 11.2.0 of CPExpert(tm)
(C) Copyright 2001, Computer Management Sciences, Inc.
PAGE: 1 - PART 1
THE FOLLOWING GUIDANCE WAS PROVIDED TO THE WLM
COMPONENT DURING THIS EXECUTION OF CPEXPERT.
SELECTSW = N * SELECT SPECIFIC SERVICE CLASSES TO ANALYZE
SELECT1 = TPNSEVEN * ANALYZE TPNSEVEN SERVICE CLASS
PERIOD1 = 1 * ANALYZE PERIOD 1 OF TPNSEVEN SERVICE CLASS
EXCLUDSW = N * EXCLUDE SPECIFIC SERVICE CLASSES
EXCLUDE1 = CICSRGN * EXCLUDE TMNSODD SERVICE CLASS
AVGRESP = ELAPSED * COMPUTE RESPONSE TIME BASED ON ELAPSED TIME
PHASE = EXECUTION * ANALYZE WORK MANAGER EXECUTION PHASE
MINTRANS = 10 * IGNORE INTERVALS WITH LESS THAN 10 TRANS.
MINSAMP = 100 * IGNORE SRV CLASS WITH LESS THAN 100 SAMPLES
HIGHCPU = 75 * PERCENT TO USE IN REPORTING HIGH CPU USERS
MAXVEL = 30
MAXRESP = 0:05:00
WLMSIG = 0.100 * SIGNIFICANT PERCENT DEGRADATION
PERFINDX = 1.1 * PERFORMANCE INDEX-MISSED GOALS EXCEED THIS
CHKPLCY = Y * CHECK SERVICE POLICY
WLMDATES = 01FEB1991 * START DATE FOR DATA ANALYSIS
WLMTIMES = 08:00:00 * START TIME FOR DATA ANALYSIS
WLMDATEE = 20MAR2001 * END DATE FOR DATA ANALYSIS
WLMTIMEE = 17:59:59 * END TIME FOR DATA ANALYSIS
WLMDAT2S = 0 * DEFAULT SECOND SELECTION DATE - START
WLMTIM2S = 0 * DEFAULT SECOND SELECTION TIME - START
WLMDAT2E = 0 * DEFAULT SECOND SELECTION DATE - END
WLMTIM2E = 0 * DEFAULT SECOND SELECTION TIME - END
SHIFT = Y * START AND END TIMES APPLY TO A SHIFT
VERBOSE= = V= * RESULTS: VERBOSE/CONCISE/SUMMARY/HISTORY
SYSTEM= = SYS1= * SPECIFY SYSTEM TO PROCESS (*ALL = ALL)
LISTRPT= = N=
HTML=Y; = Y=Y; * USE SAS ODS FEATURES
HTMLLINK = Y * LINK TO CPEXPERT DOCUMENTATION
FRAME = WLMFRAME.HTM * USE GENERIC ODS FRAME NAME
CONTENTS = WLMCONTENTS.HTM * USE GENERIC ODS CONTENTS NAME
BODY = WLMBODY.HTM * USE GENERIC ODS BODY NAME
PATH = * DO NOT SPECIFY PATH FOR ODS OUTPUT
COUPLE = Y
EXCLASSW = Y
PCTSML = 99
|
| Summary of analysis performed |
WORKLOAD MANAGER ANALYSIS, Tue, Mar 20, 2001 - SYSTEM: SYS1
Analyzed by Version 11.2.0 of CPExpert(tm)
(C) Copyright 2001, Computer Management Sciences, Inc.
PAGE: 2 - PART 1
SUMMARY OF WLM COMPONENT OPERATION, EXECUTED ON Fri, Jun 15, 2001
TOTAL RECORDS READ FROM MXG PDBLIB.TYPE72GO INPUT FILE: 17,132
RECORDS SELECTED BASED UPON SYSTEM/START/END CRITERIA: 1,839
REPORT CLASS RECORDS REJECTED: 1,497
SERVICE CLASS PERIOD RECORDS SELECTED (SELECT LOGIC): 342
SERVICE CLASS RECORDS EXCLUDED (EXCLUDE LOGIC): 0
RECORDS REJECTED (SYSTEM SERVICE CLASS, DISC. GOAL, SERVER, ETC.): 207
TOTAL RECORDS ANALYZED FOR POOR PERFORMANCE: 135
RECORDS WITH PERFORMANCE WORSE THAN GOAL FOR SERVICE CLASS PERIOD: 2
TOTAL RULE RECORDS WRITTEN BY THE WLM COMPONENT: 114
|
| Service policy in effect during analysis |
WORKLOAD MANAGER ANALYSIS, Tue, Mar 20, 2001 - SYSTEM: SYS1
Analyzed by Version 11.2.0 of CPExpert(tm)
(C) Copyright 2001, Computer Management Sciences, Inc.
PAGE: 3 - PART 2
SYSPLEX: PLEXPROD SERVICE POLICY: POLICY#1
POLICY EFFECTIVE: 16MAR2001:23:55:59
OPERATING SYSTEM: OS/390 Version 2, Release 10
MXG VERSION NUMBER: 18.18
SERVICE COEFICIENTS: CPU=1.0 SRB=1.0 MSO=0.0 IOC=0.5
SERVICE CLASS PERIOD GOAL GOAL CPU CPU
CLASS PERIOD DURATION GOAL TYPE IMPORTANCE GOAL PERCENT MIN MAX
P_BTCHIG 1 EX. VELOCITY 2 40%
P_BTCLOW 1 EX. VELOCITY 5 20%
P_BTCMED 1 EX. VELOCITY 4 30%
P_HRNSPN 1 EX. VELOCITY 2 40%
P_ONLHIG 1 EX. VELOCITY 2 90%
P_ONLLOW 1 EX. VELOCITY 4 60%
P_ONLMED 1 EX. VELOCITY 2 70%
P_REDLIN 1 EX. VELOCITY 2 40%
SYSSTC 1 SYSTEM TASKS 0
SYSTEM 1 SYSTEM TASKS 0
TSO 1 1200 % RESPONSE 3 0.500 90
TSO 2 2500 % RESPONSE 4 00:02.0 90
TSO 3 20000 % RESPONSE 4 00:15.0 90
The above may not be the complete definition of Service Policy POLICY#1.
If no address spaces were active in a service class, SMF Type 72 information
may not be available for the service class. Consequently, CPExpert would
be unable to determine the service class parameters.
The above policy is listed alphabetically. You can have the policy
ordered by service class period importance by changing the POLORDER
guidance variable.
|
| System: *ALL, Overview of rule results |
| System: SYS1, Overview of rule results |
WORKLOAD MANAGER ANALYSIS, Tue, Mar 20, 2001 - SYSTEM: SYS1
Analyzed by Version 11.2.0 of CPExpert(tm)
(C) Copyright 2001, Computer Management Sciences, Inc.
PAGE: 5 - PART 3
OVERVIEW OF RULE RESULTS
WLM058 WLM059 WLM102 WLM103 WLM107 WLM140 WLM170 WLM400 WLM602 WLM603
----------------------------------------------------------------------
20MAR2001:08:05:00 X X X X X
20MAR2001:08:20:00 X
20MAR2001:08:35:00 X
20MAR2001:08:50:00 X X X
20MAR2001:09:05:00 X
20MAR2001:09:20:00 X X
20MAR2001:09:35:00 X X
20MAR2001:09:50:00 X X X
20MAR2001:10:05:00 X X
20MAR2001:10:20:00 X
20MAR2001:10:35:00 X
20MAR2001:10:50:00 X
20MAR2001:11:05:00 X
20MAR2001:11:20:00 X X
20MAR2001:11:35:00 X
20MAR2001:11:50:00 X
20MAR2001:12:05:00 X
20MAR2001:12:20:00 X
20MAR2001:12:35:00 X
20MAR2001:12:50:00 X
20MAR2001:13:05:00 X
20MAR2001:13:20:00 X
20MAR2001:13:35:00 X
20MAR2001:13:50:00 X X
20MAR2001:14:05:00 X X
20MAR2001:14:20:00 X
20MAR2001:14:35:00 X
20MAR2001:14:36:14
20MAR2001:14:50:00 X X
20MAR2001:14:51:14
20MAR2001:15:05:00 X X
20MAR2001:15:06:14
20MAR2001:15:20:00 X
20MAR2001:15:21:14
20MAR2001:15:35:00 X
20MAR2001:15:36:14
20MAR2001:15:50:00 X X X X
20MAR2001:15:51:14
20MAR2001:16:05:00 X X
20MAR2001:16:06:14
20MAR2001:16:20:00 X
20MAR2001:16:21:14
20MAR2001:16:35:00 X
20MAR2001:16:36:14
20MAR2001:16:50:00 X
20MAR2001:16:51:14
20MAR2001:17:05:00 X
20MAR2001:17:06:14
20MAR2001:17:20:00 X
20MAR2001:17:21:14
20MAR2001:17:35:00 X
20MAR2001:17:36:14
|
| Service policy findings |
WORKLOAD MANAGER ANALYSIS, Fri, Mar 16, 2001 - SYSTEM: *ALL
Analyzed by Version 11.2.0 of CPExpert(tm)
(C) Copyright 2001, Computer Management Sciences, Inc.
PAGE: 6 - PART 3
RULE WLM002: CONFLICT EXISTS BETWEEN SERVICE CLASS AND REPORT CLASS .
The Service Policy being analyzed (Policy POLICY#1) contains a Report
Class which conflicts with a Service Class. This conflict could cause
double accounting of system resources in reports which key off the
class name. This finding has no known effect on performance, but could
have a HIGH IMPACT on accounting, billing, or capacity planning. The
following Report Class conflicts with a Service Class of the same name:
REPORT CLASS
TSO
RULE WLM005: VELOCITY GOAL MAY BE TOO HIGH FOR BATCH SERVICE CLASS .
CPExpert noticed that P_BTCHIG Service Class (Period 1) had an execution
velocity goal of 40. The P_BTCHIG Service Class had the word "batch"
in its Service Class Description. Consequently CPExpert assumes that
the service class consists of batch jobs. Specifying a relatively
high velocity goal of 40 for batch jobs may cause performance problems
unless the batch jobs are well-behaved. Under some circumstances, this
velocity goal could result in 40% of the system being used by batch
workload. Please refer to Rule WLM005 in the WLM Component User Manual
for a discussion of this issue.
RULE WLM005: VELOCITY GOAL MAY BE TOO HIGH FOR BATCH SERVICE CLASS .
CPExpert noticed that P_HRNSPN Service Class (Period 1) had an execution
velocity goal of 40. The P_HRNSPN Service Class had the word "batch"
in its Service Class Description. Consequently CPExpert assumes that
the service class consists of batch jobs. Specifying a relatively
high velocity goal of 40 for batch jobs may cause performance problems
unless the batch jobs are well-behaved. Under some circumstances, this
velocity goal could result in 40% of the system being used by batch
workload. Please refer to Rule WLM005 in the WLM Component User Manual
for a discussion of this issue.
RULE WLM005: VELOCITY GOAL MAY BE TOO HIGH FOR BATCH SERVICE CLASS .
CPExpert noticed that P_REDLIN Service Class (Period 1) had an execution
velocity goal of 40. The P_REDLIN Service Class had the word "batch"
in its Service Class Description. Consequently CPExpert assumes that
the service class consists of batch jobs. Specifying a relatively
high velocity goal of 40 for batch jobs may cause performance problems
unless the batch jobs are well-behaved. Under some circumstances, this
velocity goal could result in 40% of the system being used by batch
workload. Please refer to Rule WLM005 in the WLM Component User Manual
for a discussion of this issue.
|
| System SYS1, General System Findings |
WORKLOAD MANAGER ANALYSIS, Tue, Mar 20, 2001 - SYSTEM: SYS1
Analyzed by Version 11.2.0 of CPExpert(tm)
(C) Copyright 2001, Computer Management Sciences, Inc.
PAGE: 7 - PART 4
RULE WLM058: LOCAL PAGE RESPONSE WAS SIGNIFICANTLY WORSE THAN AVERAGE .
CPExpert has detected that the device response time for a local page
data set was significantly worse than the average device response time
for other local page data sets. This situation usually is caused by
overloading the path to the local page data set or by placing local page
data sets on volumes with other data sets. During the below intervals,
there was a significant imbalance of page transfer times among the local
page data sets, and at least one service class missed its performance
goal because of delays for page-in from auxiliary storage.
AVERAGE VOLUME VOLUME
PAGE XFR WITH POOR PAGES AVG PAGE
MEASUREMENT INTERVAL TIME XFR TIME XFR'D XFR TIME
9:20- 9:35, 20MAR2001 0.008 PAG103 162 0.012
11:20-11:35, 20MAR2001 0.005 PAG102 124 0.016
14:50-15:05, 20MAR2001 0.009 PAG102 139 0.014
15:50-16:05, 20MAR2001 0.013 PAG101 898 0.020
RULE WLM059: INSUFFICIENT LOCAL PAGE DATA SETS DEFINED FOR MIGRATION .
The number of local page data sets is insufficient to allow parallel
I/O operations for migration of pages from expanded storage to auxiliary
storage. If migration occurs often, defining too few local page data
sets could have a significant impact on paging performance. During the
measurement intervals shown below, migration was relatively high and at
least one service class missed its performance goal because of paging
delays from auxiliary storage. CPExpert suggests increasing the number
of local page data sets.
AVERAGE LOCAL PAGE DATA SUGGESTED
MEASUREMENT INTERVAL MIGRATION RATE SETS ALLOCATED ALLOCATION
8:50- 9:05, 20MAR2001 11.9 3 6
9:35- 9:50, 20MAR2001 20.0 3 6
9:50-10:05, 20MAR2001 10.7 3 6
10:05-10:20, 20MAR2001 5.8 3 6
13:50-14:05, 20MAR2001 4.1 3 6
14:05-14:20, 20MAR2001 2.5 3 6
15:05-15:20, 20MAR2001 17.3 3 6
|
| System SYS1, Specific findings, Service Class P_ONLMED |
WORKLOAD MANAGER ANALYSIS, Tue, Mar 20, 2001 - SYSTEM: SYS1
Analyzed by Version 11.2.0 of CPExpert(tm)
(C) Copyright 2001, Computer Management Sciences, Inc.
PAGE: 8 - PART 5
RULE WLM103: SERVICE CLASS DID NOT ACHIEVE VELOCITY GOAL .
P_ONLMED (Period 1): Service class did not achieve its velocity goal
during the measurement intervals shown below. The velocity goal was
70% execution velocity, with an importance level of 2. The '% USING'
and '%TOTAL DELAY' percentages are computed as a function of the average
address space EXECUTING time (to exclude activity and delays not under
WLM control). The 'PRIMARY,SECONDARY CAUSES OF DELAY' are computed as
a function of the execution delay samples on the local system.
------LOCAL SYSTEM--------
% % TOTAL EXEC PERF PLEX PRIMARY,SECOND
MEASUREMENT INTERVAL USING DELAY VELOC INDX PI CAUSES OF DELAY
15:50-16:05,20MAR2001 3.5 2.8 55% 1.27 1.76 AUX PAGING(38%)
RULE WLM400: PAGE-IN (AUXILIARY STORAGE) WAS MAJOR PERFORMANCE PROBLEM .
P_ONLMED (Period 1): Page-in from auxiliary storage was a major reason
the service class period missed its performance goal. Auxiliary storage
paging caused the following delays to P_ONLMED (Period 1), shown by catego
page-in. The delays are relative to the EXECUTING time of the service
class period.
PERCENT
PAGE-IN ---PCT DELAY, BY PAGE-IN CATEGORY---
MEASUREMENT INTERVAL DELAY PVT COMM XMEM VIO HIPR ESO
15:50-16:05,20MAR2001 2.4 2.3 0.0 0.1 0.0 0.0 0.0
|
| System SYS1, Specific findings, Service Class TSO |
WORKLOAD MANAGER ANALYSIS, Tue, Mar 20, 2001 - SYSTEM: SYS1
Analyzed by Version 11.2.0 of CPExpert(tm)
(C) Copyright 2001, Computer Management Sciences, Inc.
PAGE: 9 - PART 5
RULE WLM102: SERVICE CLASS DID NOT ACHIEVE PERCENTILE RESPONSE GOAL .
TSO (Period 1): Service class did not achieve its response goal
during the measurement intervals shown below. The response goal was
90.00 percent of the transactions completing within 0.500 seconds,
with an importance level of 3. The percentages with the primary/
secondary causes of delay are computed as a function of the average
address space EXECUTING time on the local system.
--------LOCAL SYSTEM------
TRANS %
TOTAL MEETING MEETING PERF PLEX PRIMARY,SECOND
MEASUREMENT INTERVAL TRANS GOAL GOAL INDX PI CAUSES OF DELAY
8:05- 8:20,20MAR2001 22 17 77.3 4.00 0.50 UNKNOWN(78%)
RULE WLM107: RESPONSE TIME DISTRIBUTION FOR SERVICE CLASS .
TSO (Period 1): Service class did not achieve its response goal
during the measurement intervals shown below. The response goal was
90.00 percent of the transactions completing within 0.500 seconds.
The below information shows the distribution of response times:
--PERCENT COMPLETIONS RELATIVE TO GOAL--
50- 90- 100- 110- 200-
TOTAL <50% 9O% 100% 110% 200% 400% >400%
MEASUREMENT INTERVAL TRANS GOAL GOAL GOAL GOAL GOAL GOAL GOAL
8:05- 8:20,20MAR2001 22 68.2 9.1 0.0 4.5 4.5 4.5 9.1
RULE WLM140: SYSPLEX PERFORMANCE INDEX WAS SIGNIFICANTLY LESS THAN LOCAL .
TSO (Period 1): The sysplex performance index for this service class
period was significantly less than the local performance index. One
implication of this is that the Workload Manager might not attempt to
improve performance of the service class period on the local system.
Please refer to the WLM Component User Manual for a discussion of how
the sysplex performance index and local performance index are used by
the Workload Manager. This finding applies to the following measurement
intervals:
PERFORMANCE INDEX
MEASUREMENT INTERVAL LOCAL SYSPLEX
8:05- 8:20,20MAR2001 4.00 0.50
WORKLOAD MANAGER ANALYSIS, Tue, Mar 20, 2001 - SYSTEM: SYS1
Analyzed by Version 11.2.0 of CPExpert(tm)
(C) Copyright 2001, Computer Management Sciences, Inc.
PAGE: 10 - PART 5
RULE WLM170: ADDRESS SPACES WERE IDLE A SIGNIFICANT PERCENT OF TIME .
TSO: The delay information presented above is based on the EXECUTION
time of the service class (the CPU Using, Execution Delay, and Unknown
Delay). These percentages show the distribution of time while some
transaction was active. However, address spaces in the TSO Service
Class were IDLE for a significant percent of their overall active time.
The below information shows the percent of total active time in which
address spaces were executing (processing transactions) or were idle,
and the average number of Workload Manager samples per transaction.
Please refer to Rule WLM170 in the WLM Component User Manual for a
discussion of the implications of this finding.
TOTAL PCT PCT AVG SAMPLES AVG SAMPLES
MEASUREMENT INTERVAL TRANS EXECUTING IDLE PER TRANS PER MINUTE
8:05- 8:20,20MAR2001 22 1.6 98.4 4.1 6.1
|
| System SYS1, Cross System Coupling Facility (XCF) findings |
WORKLOAD MANAGER ANALYSIS, Tue, Mar 20, 2001 - SYSTEM: SYS1
Analyzed by Version 11.2.0 of CPExpert(tm)
(C) Copyright 2001, Computer Management Sciences, Inc.
PAGE: 11 - PART 5
RULE WLM602: XCF MESSAGE BUFFER LENGTH MAY BE TOO SMALL .
The XCF message buffers may be too small for the DEFAULT transport class.
Unnecessary overhead is incurred when XCF must expand the buffers to fit
a message. The CLASSLEN parameter was specified as 8,124 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
8:50- 9:05,20MAR2001 SYS5 152 732 87 971
9:50-10:05,20MAR2001 SYS5 150 914 88 1,152
16:05-16:20,20MAR2001 SYS5 327 565 88 980
RULE WLM603: XCF MESSAGE BUFFER LENGTH MAY BE TOO LARGE .
The XCF message buffer length may be too large for the TCGRS transport
class. XCF will fill the message buffer space too quickly when the
specified message length is larger than most of the messages sent.
The CLASSLEN parameter was specified as 36,796 for this transport class,
and over 99% of the messages were less than this length. You should
consider decreasing the message length for this transport class or you
may wish to split the transport class, depending upon actual message
lengths. This situation is not critical, since XCF did not exhaust its
message buffer space. The finding is produced only to alert you to a
potential problem with storage allocation. You can suppress this finding
by altering the PCTSML guidance variable in USOURCE(WLMGUIDE). This
finding applies to the following RMF measurement intervals:
SENT SMALL MESSAGES MESSAGES TOTAL
MEASUREMENT INTERVAL TO MESSAGES THAT FIT TOO BIG MESSAGES
8:05- 8:20,20MAR2001 SYS3 179,322 0 0 179,322
8:20- 8:35,20MAR2001 SYS3 174,039 0 0 174,039
8:35- 8:50,20MAR2001 SYS3 191,929 0 0 191,929
8:50- 9:05,20MAR2001 SYS3 287,449 0 0 287,449
9:05- 9:20,20MAR2001 SYS3 239,773 0 0 239,773
9:20- 9:35,20MAR2001 SYS3 223,154 1 0 223,155
9:35- 9:50,20MAR2001 SYS3 217,652 0 0 217,652
9:50-10:05,20MAR2001 SYS3 300,195 0 0 300,195
10:05-10:20,20MAR2001 SYS3 266,019 0 0 266,019
10:20-10:35,20MAR2001 SYS3 272,699 0 0 272,699
10:35-10:50,20MAR2001 SYS3 282,343 0 0 282,343
10:50-11:05,20MAR2001 SYS3 314,223 0 0 314,223
11:05-11:20,20MAR2001 SYS3 219,170 0 0 219,170
11:20-11:35,20MAR2001 SYS3 199,270 0 0 199,270
11:35-11:50,20MAR2001 SYS3 143,187 0 0 143,187
11:50-12:05,20MAR2001 SYS3 223,570 0 0 223,570
12:05-12:20,20MAR2001 SYS3 146,765 0 0 146,765
12:20-12:35,20MAR2001 SYS3 170,619 0 0 170,619
12:35-12:50,20MAR2001 SYS3 164,807 0 0 164,807
12:50-13:05,20MAR2001 SYS3 218,410 0 0 218,410
WORKLOAD MANAGER ANALYSIS, Tue, Mar 20, 2001 - SYSTEM: SYS1
Analyzed by Version 11.2.0 of CPExpert(tm)
(C) Copyright 2001, Computer Management Sciences, Inc.
PAGE: 12 - PART 5
SENT SMALL MESSAGES MESSAGES TOTAL
MEASUREMENT INTERVAL TO MESSAGES THAT FIT TOO BIG MESSAGES
13:05-13:20,20MAR2001 SYS3 217,565 0 0 217,565
13:20-13:35,20MAR2001 SYS3 212,487 0 0 212,487
13:35-13:50,20MAR2001 SYS3 224,934 0 0 224,934
13:50-14:05,20MAR2001 SYS3 228,358 0 0 228,358
14:05-14:20,20MAR2001 SYS3 144,659 0 0 144,659
14:20-14:35,20MAR2001 SYS3 200,375 0 0 200,375
14:35-14:50,20MAR2001 SYS3 176,564 0 0 176,564
14:50-15:05,20MAR2001 SYS3 195,427 0 0 195,427
15:05-15:20,20MAR2001 SYS3 130,834 0 0 130,834
15:20-15:35,20MAR2001 SYS3 195,142 0 0 195,142
15:35-15:50,20MAR2001 SYS3 189,010 0 0 189,010
15:50-16:05,20MAR2001 SYS3 241,938 0 0 241,938
16:05-16:20,20MAR2001 SYS3 278,691 0 0 278,691
16:20-16:35,20MAR2001 SYS3 210,868 0 0 210,868
16:35-16:50,20MAR2001 SYS3 115,181 0 0 115,181
16:50-17:05,20MAR2001 SYS3 206,901 0 0 206,901
17:05-17:20,20MAR2001 SYS3 115,235 0 0 115,235
17:20-17:35,20MAR2001 SYS3 128,597 0 0 128,597
17:35-17:50,20MAR2001 SYS3 135,253 0 0 135,253
RULE WLM603: XCF MESSAGE BUFFER LENGTH MAY BE TOO LARGE .
The XCF message buffer length may be too large for the TCJES transport
class. XCF will fill the message buffer space too quickly when the
specified message length is larger than most of the messages sent.
The CLASSLEN parameter was specified as 8,124 for this transport class,
and over 99% of the messages were less than this length. You should
consider decreasing the message length for this transport class or you
may wish to split the transport class, depending upon actual message
lengths. This situation is not critical, since XCF did not exhaust its
message buffer space. The finding is produced only to alert you to a
potential problem with storage allocation. You can suppress this finding
by altering the PCTSML guidance variable in USOURCE(WLMGUIDE). This
finding applies to the following RMF measurement intervals:
SENT SMALL MESSAGES MESSAGES TOTAL
MEASUREMENT INTERVAL TO MESSAGES THAT FIT TOO BIG MESSAGES
8:05- 8:20,20MAR2001 SYS5 8,611 0 0 8,611
8:20- 8:35,20MAR2001 SYS5 7,060 0 0 7,060
8:35- 8:50,20MAR2001 SYS5 7,386 0 0 7,386
8:50- 9:05,20MAR2001 SYS5 22,916 2 0 22,918
9:05- 9:20,20MAR2001 SYS5 12,260 1 0 12,261
9:20- 9:35,20MAR2001 SYS5 12,505 0 0 12,505
9:35- 9:50,20MAR2001 SYS5 8,944 0 0 8,944
9:50-10:05,20MAR2001 SYS5 25,679 1 0 25,680
10:05-10:20,20MAR2001 SYS5 15,641 1 0 15,642
10:20-10:35,20MAR2001 SYS5 5,167 0 0 5,167
10:35-10:50,20MAR2001 SYS5 3,243 0 0 3,243
10:50-11:05,20MAR2001 SYS5 9,206 1 0 9,207
11:05-11:20,20MAR2001 SYS5 6,588 2 0 6,590
11:20-11:35,20MAR2001 SYS5 2,572 0 0 2,572
11:35-11:50,20MAR2001 SYS5 2,032 0 0 2,032
11:50-12:05,20MAR2001 SYS5 11,319 1 0 11,320
WORKLOAD MANAGER ANALYSIS, Tue, Mar 20, 2001 - SYSTEM: SYS1
Analyzed by Version 11.2.0 of CPExpert(tm)
(C) Copyright 2001, Computer Management Sciences, Inc.
PAGE: 13 - PART 5
SENT SMALL MESSAGES MESSAGES TOTAL
MEASUREMENT INTERVAL TO MESSAGES THAT FIT TOO BIG MESSAGES
12:05-12:20,20MAR2001 SYS5 6,517 0 0 6,517
12:20-12:35,20MAR2001 SYS5 3,402 0 0 3,402
12:35-12:50,20MAR2001 SYS5 5,132 0 0 5,132
12:50-13:05,20MAR2001 SYS5 5,520 0 0 5,520
13:05-13:20,20MAR2001 SYS5 13,529 1 0 13,530
13:20-13:35,20MAR2001 SYS5 11,058 0 0 11,058
13:35-13:50,20MAR2001 SYS5 16,836 5 0 16,841
13:50-14:05,20MAR2001 SYS5 16,339 5 0 16,344
14:05-14:20,20MAR2001 SYS5 11,933 1 0 11,934
14:20-14:35,20MAR2001 SYS5 12,569 0 0 12,569
14:35-14:50,20MAR2001 SYS5 13,532 0 0 13,532
14:50-15:05,20MAR2001 SYS5 13,552 0 0 13,552
15:05-15:20,20MAR2001 SYS5 16,574 0 0 16,574
15:20-15:35,20MAR2001 SYS5 14,450 0 0 14,450
15:35-15:50,20MAR2001 SYS5 13,016 2 0 13,018
15:50-16:05,20MAR2001 SYS5 21,473 0 0 21,473
16:05-16:20,20MAR2001 SYS5 26,041 2 0 26,043
16:20-16:35,20MAR2001 SYS5 9,974 0 0 9,974
16:35-16:50,20MAR2001 SYS5 2,689 0 0 2,689
16:50-17:05,20MAR2001 SYS5 12,594 8 0 12,602
17:05-17:20,20MAR2001 SYS5 2,786 0 0 2,786
17:20-17:35,20MAR2001 SYS5 3,529 0 0 3,529
17:35-17:50,20MAR2001 SYS5 13,166 0 0 13,166
|