Using the information in the CPU Activity report

High LPAR/MVS BUSY TIME PERC values could indicate contention for CPU. To check this, add the N+1, … N+150 percentages in the DISTRIBUTION OF IN-READY WORK UNIT QUEUE (where N is the number of online processors). This sum is the percentage of time when at least one task could not be dispatched. A value higher than 60% implies contention for CPU.

Low LPAR/MVS BUSY TIME PERC values can indicate that other bottlenecks in the system are preventing work from being processed.

An OUT READY average value of more than 1 could reflect processor storage constraints.

Start of change
Figure 1. CPU Activity Report
                                                       C P U  A C T I V I T Y   
                                                                                                                              
            z/OS V2R4                SYSTEM ID SYSF             DATE 07/03/2019            INTERVAL 14.59.999 
                                     RPT VERSION V2R4 RMF       TIME 12.14.33              CYCLE 1.000 SECONDS
CPU        3906   CPC CAPACITY  9163        SEQUENCE CODE 00000000000C5327                                                    
MODEL      776                              HIPERDISPATCH=YES                                                                 
H/W MODEL  M03    CHANGE REASON=NONE        BOOST TYPE=ZIIP    BOOST CLASS=IPL                                                
                                                                                                                              
---CPU---    ---------------- TIME % ----------------     LOG PROC      --I/O INTERRUPTS--                                    
NUM  TYPE    ONLINE    LPAR BUSY    MVS BUSY   PARKED     SHARE %       RATE     % VIA TPI                                    
 0    CP     100.00     2.84         2.83        0.00     100.0  HIGH    0.00     0.00                                        
 1    CP     100.00     0.80         0.79        0.00     100.0  HIGH    0.00     0.00                                        
 2    CP     100.00     1.04         1.03        0.00     100.0  HIGH   59.60     5.82                                        
TOTAL/AVERAGE           1.56         1.55                 300.0         59.60     5.82                                        
 3    IIP    100.00     1.28         1.28        0.00      48.5  MED                                                          
 4    IIP    100.00     0.07         0.07        0.00       0.0  LOW                                                          
 6    IIP    100.00     0.00        -----      100.00       0.0  LOW                                                          
TOTAL/AVERAGE           0.45         0.67                  48.5

                                                       C P U  A C T I V I T Y                                                 
                                                                                                                          
            z/OS V2R4                SYSTEM ID SYSF             DATE 07/03/2019            INTERVAL 14.59.999                 
                                     RPT VERSION V2R4 RMF       TIME 12.14.33              CYCLE 1.000 SECONDS                
SYSTEM ADDRESS SPACE AND WORK UNIT ANALYSIS                                                                                   
  ---------NUMBER OF ADDRESS SPACES---------     -----------------------DISTRIBUTION OF IN-READY WORK UNIT QUEUE--------------
  QUEUE TYPES          MIN      MAX      AVG      NUMBER OF              0    10   20   30   40   50   60   70   80   90   100
                                                  WORK UNITS     (%)     !....!....!....!....!....!....!....!....!....!....!  
  IN                    73       77     73.8                                                                                  
  IN READY               0        2      0.0     <=  N         100.0     >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>  
                                                  =  N +   1     0.0                                                          
  OUT READY              0        0      0.0      =  N +   2     0.0                                                          
  OUT WAIT               0        0      0.0      =  N +   3     0.0                                                          
                                                 <=  N +   5     0.0                                                          
  LOGICAL OUT RDY        0        0      0.0     <=  N +  10     0.0                                                          
  LOGICAL OUT WAIT      53       58     56.5     <=  N +  15     0.0                                                          
                                                 <=  N +  20     0.0                                                          
  ADDRESS SPACE TYPES                            <=  N +  30     0.0                                                          
                                                 <=  N +  40     0.0                                                          
  BATCH                  0        0      0.0     <=  N +  60     0.0                                                          
  STC                  117      120    119.6     <=  N +  80     0.0                                                          
  TSO                    0        1      0.7     <=  N + 100     0.0                                                          
  ASCH                   0        0      0.0     <=  N + 120     0.0
  OMVS                  10       10     10.0     <=  N + 150     0.0                                                          
                                                 >   N + 150     0.0                                                          
  ---------NUMBER OF WORK UNITS-------------                                                                                  
  CPU TYPES            MIN      MAX      AVG     N = NUMBER OF PROCESSORS ONLINE UNPARKED (5.0 ON AVG)                        
  CP                     0        4      0.0                                                                                  
  IIP                    0        1      0.0                                                                                  
                                                                                                                              
BLOCKED WORKLOAD ANALYSIS                                                                                                     
  OPT PARAMETERS: BLWLTRPCT (%)   0.5     PROMOTE RATE:  DEFINED       31     WAITERS FOR PROMOTE:  AVG     0.000             
                  BLWLINTHD        20                    USED (%)       0                           PEAK        0
End of change
Table 1. Fields in the CPU Activity Report
Field Heading Meaning
CPU Activity If multithreading is disabled for the reporting interval (that is, the LOADxx PROCVIEW CPU parameter is in effect), the term logical core refers to a logical processor.
CPU
MODEL
H/W MODEL

SEQUENCE CODE
The hardware description of the CPU.
CPC CAPACITY Effective processor capacity available to the central processor complex (CPC), measured in MSU/h.
CHANGE REASON Reason of the capacity change:
NONE
CPC is running at normal capacity (100 % effective capacity). No capacity change occurred.
POWERSAVE
CPC is running in power-save mode. Capacity change was initiated by the user.
MACHINE
CPC is running in cycle-steering mode. Capacity change was initiated by the machine.
N/A
No capacity change reason is reported by the machine.
HIPERDISPATCH HiperDispatch mode:
YES
Active
NO
Not active
N/A
Not supported by the hardware

If the mode changed during the reporting interval, an '*' is appended (for example: NO* indicates a switch from YES to NO).

Start of changeBOOST TYPEEnd of change Start of change
The boost type that was active at the end of the interval:
NONE
Boost is inactive.
ZIIP
zIIP boost.
SPEED
Speed boost.
ALL
zIIP and speed boost are both active.
End of change
Start of changeBOOST CLASSEnd of change Start of changeStart of changeThe boost class (or period) within which boost was active:
NONE
Boost is inactive.
IPL
IPL (Startup) boost.
SHUTDOWN
Shutdown boost.
RECOVERY
Recovery Process boost.
End of changeEnd of change
CPU NUM/TYPE The logical core identification and the processor type.
TIME % ONLINE The percentage of time the logical core was online.
TIME % LPAR BUSY
The percentage of the online time that the logical core was dispatched by LPAR.
  • For a dedicated partition:
    When LOADxx PROCVIEW CORE is in effect on hardware that supports multithreading:
                       MT Core LPAR Busy Time
    LPAR BUSY TIME % = ----------------------- * 100
                            Online Time
    Otherwise:
                       Online Time - Wait Time
    LPAR BUSY TIME % = ----------------------- * 100
                            Online Time
  • For a non-dedicated partition when Wait Completion is NO:
                       Partition Dispatch Time
    LPAR BUSY TIME % = ----------------------- * 100
                             Online Time

    The partition dispatch time is the elapsed time that PR/SM dispatched this logical core during the interval.

  • For a non-dedicated partition when Wait Completion is YES:
                       Partition Dispatch Time - Wait Time
    LPAR BUSY TIME % = ----------------------------------- * 100
                                   Online Time
TIME % MVS BUSY The percentage of the online time that the logical processor was busy.
                  Online Time - (Wait Time + Parked Time)
MVS BUSY TIME % = --------------------------------------- * 100
                         Online Time - Parked Time

The MVS view of CPU time is not meaningful if the logical processor is parked during the entire reporting interval. In this case, '----' is shown.

TIME % PARKED The percentage of time that the logical processor was parked. In HiperDispatch mode, processors with a low amount of physical processor share can be parked. That is, they are not dispatched by z/OS and do not attempt to run work. Without HiperDispatch, processors are not parked and '----' is shown.
MT % PROD The percentage of the maximum core capacity that was used in the reporting interval while the logical core was dispatched to physical hardware.

When MT % PROD equals 100% and the LOADxx PROCVIEW CORE parameter is in effect, all threads on the core are executing work and all core resources are being used. If MT % PROD is less than 100%, the core resources were dispatched to physical hardware but one or more threads on a logical core were in a wait because they had no work to run.

If a core was reconfigured offline/online during the reporting interval, no multithreading core productivity is calculated and '------' is shown. If the LOADxx PROCVIEW CPU parameter is in effect, this field is not displayed.

MT % UTIL The percentage of the maximum core capacity that was used in the reporting interval.

MT % UTIL = MT Core Productivity * TIME % LPAR BUSY

If a core was reconfigured offline/online during the reporting interval, no multithreading core utilization is calculated and '------' is shown. If the LOADxx PROCVIEW CPU parameter is in effect, this field is not displayed.

LOG PROC SHARE % Percentage of the physical processor that the logical processor is entitled to use.

Without HiperDispatch, the processing weight is equally divided between the online logical processors.

In HiperDispatch mode, logical processors have a high, medium or low share of the physical processor. The share percentage is the average value for the reporting interval, whereas HIGH, MED or LOW indicates the HiperDispatch priority at the end of the reporting interval. When the priority changed during the interval, an '*' is appended.

N/A is displayed if the HiperDispatch priority is not indicated by the hardware at the end of the reporting interval.

I/O INTERRUPTS RATE The total rate per second that this processor handled I/O interrupts. The rate reflects the processing for the entire interval. This might include periods of time when the SRM enabled or disabled this processor for I/O interrupts. The rate includes interrupts handled by the second level interrupt handler (SLIH), as well as those handled by the Test Pending Interrupt (TPI) instruction.
       SLIH + TPI
RATE = ----------
          INT
SLIH
Interruptions that the second level interrupt handler handled
TPI
Interruptions that the Test Pending Interrupt instruction handled
INT
Interval time (seconds)
I/O INTERRUPTS % VIA TPI The percentage of the total interrupts for this processor during the RMF interval that are handled by the I/O supervisor without re-enabling.
                TPI
% VIA TPI = ---------- * 100
            SLIH + TPI
TPI
Interrupts that the Test Pending Interrupt instruction handled
SLIH
Interrupts that the second level interrupt handler handled
For the following three TOTAL/AVERAGE values, the logical processors that are parked during the entire interval are not considered in the calculation of the average TIME % MVS BUSY.
TOTAL/AVERAGE (CP) The average or total value for general purpose processors (standard CPs).
TOTAL/AVERAGE (zAAP) The average value for zAAPs. Only visible if zAAPs are configured online.
TOTAL/AVERAGE (zIIP) The average value for zIIPs. Only visible if zIIPs are configured online.
Multi-Threading Analysis: This information is only displayed when the LOADxx PROCVIEW CORE parameter is in effect. Multithreading information is only shown for those processor types for which at least one logical core was configured online for the complete interval.
CPU TYPE Processor type CP, IIP, or AAP.
MODE The multithreading mode of a processor type designates the number of active threads for each online logical core of this type. If MT MODE is greater than 1, multithreading becomes effective for this processor type.
MAX CF Multithreading maximum capacity factor for a processor type. The multithreading maximum capacity factor represents the ratio of the maximum amount of work that can be accomplished using all active threads to the amount of work that would have been accomplished within this reporting interval when multithreading was disabled.

'------' is shown when the multithreading maximum capacity factor cannot be calculated.

CF Multithreading capacity factor for a processor type. The multithreading capacity factor represents the ratio of the amount of work that has been accomplished within this reporting interval to the amount of work that would have been accomplished with multithreading disabled.

'------' is shown when the multithreading capacity factor cannot be calculated.

AVG TD Average thread density for a processor type. This value represents the average number of active threads for those cores that were dispatched to physical hardware.

'------' is shown when the average thread density cannot be calculated.

System Address Space and Work Unit Analysis: contains information about the NUMBER OF ADDRESS SPACES categorized by the QUEUE TYPES, in which they have been waiting, and categorized by the ADDRESS SPACE TYPES. Furthermore, the MIN, MAX, and AVG numbers of work units are categorized by the CPU TYPES for which they have been dispatched (that is, for standard CPs, zAAPs and zIIPs). This section also shows how many work units have been waiting in the IN-READY queue (DISTRIBUTION OF IN-READY WORK UNIT QUEUE).
QUEUE TYPES Shows the number of address spaces that are waiting in the different queues. For each queue type, the MIN, MAX and AVG numbers of address spaces are displayed.

The following queue types are analyzed:

IN READY
Address spaces that are in central storage and ready to execute or currently in execution.
IN
Address spaces that are in central storage (corresponds to SRM in queue). This count includes the IN READY count.
OUT READY
Address spaces on the SRM out queue that are physically swapped out of central storage and ready to execute.
Note: Some address spaces on the SRM out queue might represent those TSO/E users that the SRM intentionally delayed to meet an installation's response time objective. Because these address spaces do not represent a potential performance problem, they are not included in the value reported for OUT READY.
OUT WAIT
Address spaces on the SRM wait queue that are physically swapped out of central storage and not ready to execute.
LOGICAL OUT RDY
Address spaces on the SRM out queue that are physically in central storage but logically swapped out of central storage and ready to execute.
LOGICAL OUT WAIT
Address spaces on the SRM wait queue that are physically in central storage but logically swapped out of central storage and not ready to execute.
ADDRESS SPACE TYPES Shows the total number of address spaces detected during the report interval, categorized by address space types. For each address space type, the MIN, MAX and AVG numbers of active address spaces are displayed.

The following ADDRESS SPACE TYPES are analyzed:

BATCH
Address spaces used for batch jobs.
STC
Address spaces used for started task controls.
TSO
Address spaces used for TSO/E users.
ASCH
APPC/MVS transaction scheduler (ASCH) address spaces.
OMVS
Address spaces for z/OS UNIX System Services.
DISTRIBUTION OF IN-READY WORK UNIT QUEUE The percentaged and graphical distribution of SRM samples when the number of work units on the IN-READY queue is within a certain range. The correlation is based on N, which is the number of online logical processors when the sample is taken. In HiperDispatch mode, N is the number of online logical processors that are not parked.

For example, NUMBER OF WORK UNITS = N + 10 with a percentage of 4.3 (see Figure 1) indicates that in 4.3 % of the samples ten work units were waiting for a logical processor.

NUMBER OF WORK UNITS
by CPU type
The minimum, maximum and average numbers of running and waiting work units categorized by CPU type (standard CPs, zAAPs and zIIPs).
Blocked Workload Analysis: provides information about blocked address spaces and enclaves.
OPT
PARAMETERS
Lists the OPT parameters which define the workload promotion:
BLWLTRPCT (%)
Specifies how much of the CPU capacity is to be used to promote blocked workloads.

This parameter does not influence the amount of CPU service that a single blocked address space or enclave is given. Instead, this parameter influences how many different address spaces or enclaves can be promoted at the same point in time. If the value specified with this parameter is not large enough, blocked workloads might need to wait longer than the time interval defined by BLWLINTHD.

This value is specified as a number between 0 and 200 where 200 accounts for 20.0%.

BLWLINTHD
Specifies the threshold time interval in seconds for which a swapped-in address space or enclave must wait before being considered to be blocked and eligible for promotion.

If the parameters have been changed during the reporting interval, the values are followed by an '*'.

PROMOTE RATE
DEFINED
Number of blocked dispatchable work units which may get promoted in their dispatching priority per second. This value is derived from OPT parameter BLWLTRPCT.
USED (%)
The utilization of the defined promote rate during the reporting interval.
WAITERS FOR PROMOTE Number of address spaces and enclaves found blocked according to OPT parameter BLWLINTHD:
AVG
Average number found blocked during the report interval.
PEAK
Highest number found blocked during the report interval.