topasout Command

Purpose

Generates reports by processing xmwlm, nmon, and topas recordings.

Syntax

Local reports

topasout -R type [-i interval ] [ -b time ] [-e time ] topas_recording_file

Comma-separated report

topasout -c [ -m type] topas_recording_file

Spread-sheet report

topasout [ -s ] [ -m type] topas_recording_file

Nmon analyzer report

topasout -a topas_recording_file

WLE Report from topasrec / nmon file

topasout -R wle { nmon_recording_file | topas_recording_file }

CEC reports

topasout -R type [ -i interval ] [ -b time ] [ -e time ] topas_recording_file

Comma-separated report

topasout [ -c ] topas_recording_file

Spread-sheet report

topasout -s topas_recording_file

Description

The topasout command is used to convert the binary recordings generated by the xmwlm, xmtrend, or topasrec utilities. The binary recording can be the local system recording, the central electronic complex (CEC) recording, or the cluster recording. Through SMIT, you can enable, configure, or disable a binary recording.

If there is more than one value for a metric within the user-specified interval, the topasout command averages out all of the values to get single value that can be printed in the report. For values that cannot be averaged out (like simultaneous multithreading, dedicated and shared modes), the topasout command takes the last or the first values that are recorded in the interval.

Local reports

There are several types of local reports: the Summary report, the Detailed report, the LAN report, the Disk report, the Comma-separated report, the Nmon analyzer report, the Adapter report, and the Virtual adapter report.

Summary report

A Summary report presents the consolidated view of system information.

The following column headings are in a summary report:

Item Description
Time Ending time of the report interval. Metric values are averaged out over this interval and printed in the report
InU Memory that is used
Us Percentage of processor time spent in the user mode
Sy Percentage of processor time spent in the system mode
Wa Percentage of processor time spent waiting for I/O
Id Percentage of time that the processor is idle
PhysB Percentage of physical processors that are busy
RunQ The average number of threads that are ready to run but are waiting for a processor to become available
WtQ The average number of threads that are waiting for paging to be completed
Cswitch The number of context switches per second in the reporting interval
Syscall The number of system calls executed per second in the reporting interval
PgFault The number of I/O and other page faults
%don Sum of %idle cycles donated and %busy cycles donated
%stl Sum of %idle cycles stolen and %busy cycles stolen
The following sample shows the output of a local Summary report:
Report: System Summary  --- hostname: aixfvt19                      version:1.1
Start:01/24/07 04:45:50   Stop:01/24/07 04:48:07  Int: 5 Min  Range:   2 Min
Mem:  1.2 GB  Dedicated SMT: ON   Logical CPUs:  2
Time       InU Us Sy Wa  Id  PhysB RunQ  WtQ CSwitch Syscall PgFault 
04:48:07   1.2  3  0  0  88   3.43  1.1  0.0     168     893      23

Detailed report

A detailed report provides a detailed view of the system metrics.

The following column headings are in a detailed report:
Item Description
Mode The information about the following modes are reported:
  • Don represents donating dedicated partition.
  • Ded represents that the dedicated partition are not donating or the donation is not enabled.
  • Shr represents shared mode.
Lp Number of logical processors.
SMT Status of the SMIT. It is On when the SMT is enable. It is Off when the SMT is disabled.
Ent Entitlement granted (shared-only).
Poolid Pool ID. This column is applicable only if this partition belongs to a valid shared processor pool.
Kern Percentage of processor time spent in the kernel mode.
User Percentage of processor time spent in the user mode.
Wait Percentage of processor time spent for waiting for I/O.
Idle Percentage of time that the processor is idle.
PhysB Percentage of physical processors that are busy.
Entc Percentage of entitled capacity that is consumed. This heading is applicable for shared partition only.
Sz, GB (in Memory section) Memory size in gigabytes.
InU (in Memory section) Memory used in gigabytes.
%Comp Percentage of real memory that is allocated to computational page frames. Computational page frames are backed by paging space.
%Nonc Percentage of real memory that is allocated to non-computational page frames. Non-computational page frames are backed by file space: either data files, executable files, or shared library files.
%Clnt Percentage of real memory that is allocated to cache, remotely mounted files.
Sz, GB (in Paging section) Paging space in gigabytes.
InU (in Paging section) Paging space used in gigabytes.
Flt Total number of page faults that are taken per second in the reporting interval. This includes page faults that do not cause paging activity.
Pg-I Number of 4 K pages that are read per second in the reporting interval.
Pg-O Number of 4 K pages that are written per second in the reporting interval.
Bdon Percentage of physical processor that is used while busy cycles are being donated. This metric is applicable only for donating dedicated partitions.
Idon Percentage of physical processor that is used while explicitly donating idle cycles. This metric is applicable only for donating dedicated partitions.
Istl Percentage of physical processor that is used while idle cycles are being stolen by the hypervisor. This metric is applicable only for dedicated partitions.
Bstl Percentage of physical processor that is used while busy cycles are being stolen by the hypervisor. This metric is applicable only for dedicated partitions. The %idon and %bdon metrics are not displayed when no dedicated partition is donating.
Vcsw Average number of virtual context switches per second in the reporting interval.
Phint Average number of phantom interrupts per second in the reporting interval. This column is applicable only to shared partitions.
Cswth Number of process context switches per second in the reporting interval.
Syscl Number of system calls per second run in the reporting interval.
RunQ Average number of threads that are ready to run but are waiting for a processor to become available.
WtQ Average number of threads that are waiting for paging to complete.
SrvV2 Number of NFS Server V2 calls per second in the reporting interval.
CltV2 Number of NFS Client V2 calls per second in the reporting interval.
SrvV3 Number of Server V3 calls per second in the reporting interval.
CltV3 Number of Client V3 calls per second in the reporting interval.
Network Name of the network interface.
I-Pack Number of data packets that are received per second.
O-Pack Number of data packets that are sent per second in the reporting interval.
KB-I Number of kilobytes that are received per second in the reporting interval.
KB-O Number of kilobytes that are sent per second in the reporting interval.
Disk Name of the physical disk.
Busy% Percentage of time that the physical disks are active (bandwidth utilization for the drive).
KBPS Number of kilobytes that are read and written per second in the reporting interval. This column is the sum of the KB-R and KB-W metrics.
TPS Number of transfers per second that are issued to the physical disk. A transfer is an I/O request to the physical disk. Multiple logical requests can be combined into a single I/O request to the disk. The size of a transfer is not determinate.
KB-R Number of kilobytes that are read per second from the physical disk in the reporting interval.
KB-W Number of kilobytes that are written per second to the physical disk in the reporting interval.
The following sample shows a local Detailed report:
Sample output
#Report: System Detailed --- hostname: ptoolsl1                  version: 1.2
Start:12/21/05 10.00.00 Stop:12/21/05 11.00.00  Int: 5 Min  Range: 60 Min 
Time: 10.00.00 -------------------------------------------------------
CONFIG        CPU             MEMORY        PAGING      
Mode   Don    Kern   12.0     Sz,GB  16.0   Sz,GB  4.0
LP     4      User    8.0     InU     4.3   InU    2.3
SMT    ON     Wait    0.0     %Comp   3.1   Flt    221
Ent    3.0    Idle   80.0     %NonC   9.0   Pg-I    87
Poolid 3      PhyB    0.7     %Clnt   2.0   Pg-O    44
              EntC    8.0

PHYP          EVENTS/QUEUES   NFS
Bdon    0.1   Cswth    3213   SrvV2    32
Idon    0.5   Syscl   43831   CltV2    12
Bstl    0.5   RunQ        1   SrvV3    44
Istl    0.4   WtQ         0   CltV3    18
Vcsw    1214
Phint   120                          

Network  KBPS   I-Pack  O-Pack    KB-I    KB-O
en0       0.6      7.5     0.5     0.3     0.3
en1      22.3    820.1   124.3   410.0    61.2
lo0       0.0      0.0     0.0     0.0     0.0

Disk    Busy%     KBPS     TPS    KB-R    KB-W  
hdisk0    0.0      0.0     0.0     0.0     0.0   
hdisk1    0.0      0.0     0.0     0.0     0.0

		topasout local report – detailed report

Disk reports

A Disk report provides information about the amount of data that are read or written to disks.

The following column headings are in a Disk report:
Item Description
Mem Total memory that are available in gigabytes at the first reporting interval.
Logical CPUs Number of logical processors at the first reporting interval.
Time Ending time of the reporting interval. Metric values are averaged out over this time interval and printed in the report.
InU Total memory that are used in gigabytes.
PhysB Percentage of physical processors that are busy.
MBPS Number of megabytes that are read and written per second. This column is the sum of the MB-W and MB-R metrics.
TPS Number of transfers per second that are issued to the physical disk. A transfer is an I/O request to the physical disk. Multiple logical requests can be combined into a single I/O request to the disk. The size of a transfer is not fixed.
MB-R Data that are read in megabytes per second from the physical disk.
MB-W Data that are written in megabytes per second to the physical disk.
The following sample shows the output of a local Disk report:
Sample output
Report: Total Disk I/O Summary  --- hostname: aixfvt19   version:1.1
Start:01/24/07 04:45:50   Stop:01/24/07 04:48:07  Int: 5 Min  Range:15 Min
Mem:  1.2 GB   Dedicated SMT: ON   Logical CPUs:  2
Time      InU   PhysB   MBPS    TPS   MB-R   MB-W
04:48:07  1.2     3.4    0.2    2.1    0.1    0.1
04:53:07  1.2     3.4    0.3    2.1    0.0    0.3
...

LAN reports

A LAN report provides the amount of data that are received or sent in the network interfaces.

The following column headings are in a LAN report:
Item Description
Mem Total memory available in gigabytes at the first reporting interval.
Logical processors Number of logical processors at the first reporting interval.
Time Ending time of the reporting interval. Metric values are averaged out over this time interval and printed in the report.
InU Total memory used in gigabytes.
PhysB Percentage of physical processors that are busy.
MBPS Sum of the MB-I and MB-O values. It equals to the data in megabytes sent and received per second.
MB-I Data in megabytes that are received per second in the reporting interval.
MB-O Data in megabytes that are sent per second in the reporting interval.
Xmtdrp Average amount of transmitted packets that are dropped per second at device driver level in the reporting interval.
Rcvdrp Average amount of received packets that are dropped per second at device driver level in the reporting interval.
The following sample shows the output of a local LAN report:
#Report: System LAN Summary  --- hostname: tooltime2                     version:1.1
Start:03/02/07 00:38:18 Stop:03/02/07 07:08:32   Int:  5 Min   Range:  390 Min
Mem:  4.0 GB  Shared SMT: ON  Logical CPUs:  2
Time       InU   PhysB   MBPS  MB-I  MB-O Rcvdrp Xmtdrp
00:43:18   0.6     0.1    0.0   0.0   0.0      0      0
00:48:18   0.6     0.3    0.0   0.0   0.0      0      0
00:53:19   0.7     0.2    0.0   0.0   0.0      0      0
...

Nmon analyzer style output

The topasout command generates a Nmon analyzer report that can be viewed with the nmon analyzer.

The topasout command is used to post process the binary recordings generated by the xmwlm utility the xmtrend utility and thetopasrec utility. The binary recording can be the Local System recording, Central Electronic Complex (CEC) recording or Cluster recording. Through SMIT you can enable, configure or disable a binary recording.

Note: The xmwlm and xmtrend utilities are obsolete and are replaced by the topasrec utility.
Use the topasout command with the -a flag to generate this report. You can open the generated .csv file with a nmon analyzer. For example, to generate a xmwlm.061016.csv file, enter the following command:
topasout -a /etc/perf/daily/xmwlm.061016
The generated .csv file locates in the same directory of the original file, that is, in the /etc/perf/daily/ directory. The file name is xmwlm.061016.csv.

Comma-separated report

The topasout command generates a report that contains data that is separated with comma.

Use the topasout command with the -c flag to generate this report. The output file is written to recordedfilename_01 file.

For example, to generate a comma-separated report for the xmwlm.060503 file, enter the following command:
topasout -c /etc/perf/daily/xmwlm.060503

The output file is the xmwlm.060503_01 file which locates in the same directory as the original file.

When you specify the -m flag, the topasout command writes the min, max, mean, stdev, and the exp values of the recorded metrics in the report.

The following sample shows the output of a local report with the data separated by commas:
#Monitor: xmtrend recording--- hostname: aixfvt19 ValueType: mean
Time="2007/01/24 04:45:50", CPU/gluser=0.02
Time="2007/01/24 04:45:50", CPU/glkern=0.28
Time="2007/01/24 04:45:50", CPU/glwait=0.00
Time="2007/01/24 04:45:50", CPU/glidle=99.69
Time="2007/01/24 04:45:50", NFS/Server/v3calls=0.00
Time="2007/01/24 04:45:50", NFS/Server/v2calls=0.00
…

Spreadsheet format report

The topasout command generates a report in spreadsheet format.

Use the topasout command with -s flag to generate this report. The output file is written to recordedfilename_01 file.

For example, to generate a report in spreadsheet format for the xmwlm.060503 file, enter the following command:
topasout -s /etc/perf/daily/xmwlm.060503

The output file is the xmwlm.060503_01 file which locates in the same directory as the original file.

When you specify the -m flag, the topasout command writes the min, max, mean, stdev, and the exp values of the recorded metrics in the report.

Adapter report

An Adapter report provides information about the amount of data that is read or written to adapters.

The following metrics of the adapter are in the report:
Item Description
Adapter Name of the adapter
KBPS Amount of data transferred (read or written) in the adapter in kilobytes per second
TPS Number of transfers per second that are issued to the adapter
KB-R Number of kilobytes read from the adapter
KB-W Number of kilobytes written to the adapter
The following metrics of the disks are in the report:
Item Description
Vtargets/Disk Name of the virtual target device or disk.
Busy% Percentage of time that the virtual target device or disk is active (bandwidth use for the drive).
KBPS Number of kilobytes read and written per second over the monitoring interval. This field is the sum of the values of the KB-R and KB-W metrics.
TPS Number of transfers per second that are issued to the virtual target device or disk. A transfer is an I/O request to the virtual target device or disk. Multiple logical requests can be combined into a single I/O request to the disk. A transfer is of medium size.
KB-R Number of kilobytes read per second from the virtual target device or disk.
KB-W Number of kilobytes written per second to the virtual target device or disk.
AQD Average number of requests waiting to be sent to the virtual target device or disk.
AQW Average queue that is waiting per request reported in millisecond. The suffix indicates the unit of time. The default time unit is millisecond.
ART Average time to receive a response from the hosting server for the read request sent. The suffix indicates the unit of time. The default time unit is millisecond.
AWT Average time to receive a response from the hosting server for the write request sent. The suffix indicates the unit of time. The default time unit is millisecond.
MRT Maximum time to receive a response from the hosting server for the read request sent. The suffix indicates the unit of time. The default time unit is millisecond.
MWT Maximum time to receive a response from the hosting server for the write request sent. The suffix indicates the unit of time. The default time unit is millisecond.

Virtual adapter report

The following metrics of the adapter are reported in the Virtual adapter report:
Item Description
vAdapter Name of the adapter.
KBPS Amount of data transferred (read or written) in the adapter in kilobytes per second.
TPS Number of transfers per second that are issued to the adapter.
KB-R Number of blocks received per second from the hosting server to this adapter.
KB-W Number of blocks sent per second from this adapter to the hosting server.
AQD Number of requests waiting to be sent to adapter.
AQW Time spent by a transfer request in the wait queue. Reported in millisecond. The suffix indicates the unit of time. The default time unit is millisecond.
ART Time to receive a response from the hosting server for the read request sent The suffix indicates the unit of time. The default time unit is millisecond.
AWT Time to receive a response from the hosting server for the write request sent The suffix indicates the unit of time. The default time unit is millisecond.
MRT Maximum time to receive a response from the hosting server for the read request sent. The suffix indicates the unit of time. The default time unit is millisecond.
MWT Maximum time to receive a response from the hosting server for the write request sent. The suffix indicates the unit of time. The default time unit is millisecond.
The following metrics of the disks are in the report:
Item Description
Vtargets/Disk Name of the virtual target device or disk.
Busy% Percentage of time that the virtual target device or disk is active (bandwidth use for the drive).
KBPS Number of kilobytes read and written per second over the monitoring interval. This field is the sum of the values of the KB-R and KB-W metrics.
TPS Number of transfers per second that are issued to the virtual target device or disk. A transfer is an I/O request to the virtual target device or disk. Multiple logical requests can be combined into a single I/O request to the disk. A transfer is of medium size.
KB-R Number of kilobytes read per second from the virtual target device or disk.
KB-W Number of kilobytes written per second to the virtual target device or disk.
AQD Average number of requests waiting to be sent to the virtual target device or disk.
AQW Average queue that is waiting per request reported in milliseconds. The suffix indicates the unit of time. The default time unit is millisecond.
ART Average time to receive a response from the hosting server for the read request sent. The suffix indicates the unit of time. The default time unit is millisecond.
AWT Average time to receive a response from the hosting server for the write request sent. The suffix indicates the unit of time. The default time unit is millisecond.
MRT Maximum time to receive a response from the hosting server for the read request sent. The suffix indicates the unit of time. The default time unit is millisecond.
MWT Maximum time to receive a response from the hosting server for the write request sent. The suffix indicates the unit of time. The default time unit is millisecond.

On Demand WLE input from topasrec / nmon recordings

In addition to weekly peak inputs for WLE through SMIT, the user can invoke an On Demand WLE input file to study a particular workload and use that data to size the systems and generate reports. The topasout command has the capability to study a particular topas or nmon recording and generate WLE readable reports in xml format using this option.

Use topasout -R wle <filename> option to generate the WLE report. For example, to generate a report from file, use the following command.

topasout –R wle /etc/perf/daily/xmwlm.060503

The wle option is different from the other types of –R in a way that both topas and nmon recordings can be given as an input to this option while only topas recordings (recordings generated through xmwlm and topasrec) can be given as an input file for the other options.

CEC reports

There are five types of CEC reports: the Summary report, the Detailed report, the Shared processor pool report, the Comma-separated report, and the Spread-sheet report.

Summary report

This report provides a summary of the CEC system. The reporting is based on the partitions that actually responded to the topas command. If the partitions in the CEC do not have the xmtopas or xmservd configured, the partitions cannot be monitored.

A CEC summary report contains the following column headings:

Header (partition details):
Item Description
Mon Number of the partitions that are monitored in the first reporting time interval
UnM Number of the partitions that are not monitored in the first reporting time interval
Shr Number of the shared partitions in the first reporting time interval
Ded Number of the dedicated partitions in the first reporting time interval
Cap Number of the capped partitions in the first reporting time interval
UnC Number of the uncapped partitions in the first reporting time interval

CEC:

Item Description
ShrB Shared physical processor busy. (Sum of physical busy of processors in the shared partitions.)
DedB Dedicated physical processor busy. (Sum of physical busy of processors in the dedicated partitions.)
Don Total number of the processors that are donated to the physical pool.

Processors:

Item Description
Mon Number of the physical processors that are monitored
UnMon Number of the physical processors that are not monitored.
Shr Number of the processors in shared partitions
Ded Number of the processors in dedicated partitions
PSz Number of the active shared processors in the physical pool
APP Available physical processors in the pool
Memory (GB):
Item Description
Mon Total memory of the monitored partitions
UnM Total memory of the partitions that are not monitored
Avl Memory available to partitions
InUse Memory in used in the monitored partitions
UnA Memory that is not available for partitions
The following sample shows the output of a CEC Summary report:
Sample Output
#Report: CEC Summary  --- hostname: ptoolsl3                   version:1.2
Start:02/22/07 00:44:06   Stop:02/22/07 23:59:06  Int: 5 Min  Range:1395 Min
Partition Mon:  3  UnM:  0  Shr:  1  Ded:  2  Cap:  2  UnC:  1
       -CEC----------------  -Processors---------------- -Memory (GB)-----------
Time   ShrB  DedB  Don  Stl  Mon  UnM  Shr Ded  PSz  APP Mon  UnM  Avl  UnA InU
00:49  0.00  0.00   -    -   2.2  0.0  0.2   2  2.0  2.0 9.4  0.0  8.0  0.0 1.0
00:54  0.00  0.00   -    -   2.2  0.0  0.2   2  2.0  2.0 9.4  0.0  8.0  0.0 1.0
00:59  0.00  0.00   -    -   2.2  0.0  0.2   2  2.0  2.0 9.4  0.0  8.0  0.0 1.0

Detailed report

A CEC Detailed report gives a detailed view of all the partitions that the topas command is able to record data from.

The followings column headings are in a CEC Detailed report:

Partition Info:
Item Description
Monitored Number of partitions that are monitored
Unmonitored Number of partitions that are not monitored
Shared Number of shared partitions
Uncapped Number of uncapped shared partitions
Capped Number of capped shared partitions
Dedicated Number of dedicated partitions
Donating Number of partitions that are donating

Memory:

Item Description
Monitored Total memory that is monitored
UnMonitored Total memory that is not monitored
Available Total memory that is available
UnAllocated Total memory that is not allocated to any partition
Consumed Total memory that is consumed by the partitions

Processor:

Item Description
Monitored Number of physical processors that are monitored
UnMonitored Number of physical processors that are not monitored
Available Number of physical processors that are available in the CEC system
UnAllocated Number of physical processors that are not allocated to any partition
Shared Number of processors in shared partitions
Dedicated Number of processors in dedicated partitions
Donated Sum of the number of processors in all of the partitions that are currently donating
Pool Size Number of active shared processors in the physical pool
Avail Proc Pool Available physical processors in pool. This is the idle cycles in the pool reported as a number of processors
Shr Physical Busy Sum of the busy physical processors of all of the shared partitions
Ded Physical CPUs Sum of the busy physical processors of all of the dedicated partitions
Donated Phys. CPUs Sum of the donated processor cycles (reported as a number of processors) from all partitions
Stolen Phys. CPUs Sum of the stolen processor cycles (reported as a number of processors) from all partitions
Virtual Pools Number of the virtual pools
Virt. Context Switch Total number of the virtual context switches per second in the monitoring interval
Phantom Interrupts Total number of the phantom interrupts per second in the monitoring interval

Individual partition data:

Item Description
Host Host name
OS Operating system level
M The M column heading represents the mode.
In shared partitions, it displays the following attributes:
  • C- SMT is enabled and capped
  • c- SMT is disabled and capped
  • U- SMT is enabled and uncapped
  • u- SMT is disabled and uncapped
In dedicated partitions, it displays the following attributes:
  • S- SMT is enabled and is not donating
  • d- SMT is disabled and donating
  • D- SMT is enabled and donating
Mem Total memory in gigabytes
InU Memory in used in gigabytes
Lp Number of logical processors
Us Percentage of processor that is used by programs executing in the user mode
Sy Percentage of processor that is used by programs executing in kernel mode
Wa Percentage of time that is spent waiting for I/O
Id Percentage of time that the processor is idle
PhysB Number of physical processors that are busy
Ent Entitlement granted (shared only)
%Entc Percentage of entitlement consumed (shared only)
Vcsw Virtual context switches average per second (shared only)
PhI Phantom interrupts average per second (shared only)
%idon Percentage of physical processor that is used while explicitly donating idle cycles. This metric is applicable only for donating dedicated partitions.
%bdon Percentage of physical processor that is used while busy cycles are being donated. This metric is applicable only for donating dedicated partitions
%istl Percentage of physical processor that is used while busy cycles are being stolen by the hypervisor. This metric is applicable only for dedicated partitions
The following sample shows the output of a CEC Detailed report:
#Report: CEC Detailed --- hostname: ptoolsl3                   version:1.2
Start:03/06/07 07:19:39   Stop:03/06/07 07:28:39  Int: 5 Min  Range:   9 Min

Time: 07:24:38 -----------------------------------------------------------------
Partition Info   Memory (GB)        Processors         Avail Pool :  2.0
Monitored  :  3  Monitored  :  9.4  Monitored  :  2.2  Shr Physcl Busy:  0.01
UnMonitored:  0  UnMonitored:  0.0  UnMonitored:  0.0  Ded Physcl Busy:  0.01
Shared     :  1  Available  :  0.0  Available  :  0.0  Donated Phys. CPUs: 0.00
UnCapped   :  1  UnAllocated:  0.0  Unallocated:  0.0  Stolen Phys. CPUs : 0.00
Capped     :  2  Consumed   :  0.0  Shared     :  0.2  Hypervisor
Dedicated  :  2                     Dedicated  :  2.0  Virt Cntxt Swtch:   545
Donating   :  0                     Donated    :  0    Phantom Intrpt  :     0
                                    Pool Size  :  2.0

Host         OS  M  Mem  InU Lp Us Sy Wa Id PhysB  Vcsw  Ent  %EntC PhI
-------------------------------------shared-------------------------------------
ptoolsl1     A53 U  3.1  1.9  4  0  1  0 98  0.01   317  0.2   2.55   0

Host         OS  M  Mem  InU Lp Us Sy Wa Id PhysB  Vcsw  %istl %bstl
------------------------------------dedicated-----------------------------------
ptoolsl3     A54    3.1  0.9  2  0  0  0 99  0.00   228    -     -
ptoolsl1     A52    3.1  2.7  1  0  1  0 99  0.01     0    -     -

Time: 07:28:39 ----------------------------------------------------------------

Shared-processor-pool report

The CEC Shared-processor-pool report contains information about the shared processor pools.

The following column headings are included in a Shared-processor-pool report:
Item Description
psize Effective maximum capacity of the pool.
entc Entitled capacity of the pool.
maxc Maximum capacity of the pool.
physb Sum of the physical busy of processors in the shared partitions of a pool. (The "physical busy" refers to fraction of physical processors that are busy.)
app Available physical processors in the pool.
mem Sum of the monitored memory for all of the shared partitions in the pool.
muse Sum of the memory consumed by all of the shared partitions in the pool.
The following sample shows the output of a CEC Shared-processor-pool report:
Sample Output

#Report: Topas CEC Pool Detailed --- hostname: ptoolsl1                version: 1.0
pool  psize entc  maxc physb app   mem  muse
0     3.0   2.0   3.0  0.1   1.0   2.0  1.0
1     4.0   3.0   5.0  0.5   1.5   1.0  0.5
2     3.0   2.5   4.0  0.2   2.0   1.0  0.5

Host       Pi OS  M Mem InU Lp  Us Sy Wa Id  PhysB Vcsw Ent  %EntC PhI 
--------------------------------shared------------------------------------------ 
ptools1    0  53 U  11   9  2  11  13  0 75  0.10  121 0.25   0.3   3 

ptools5    1  53 U  12  10  2  12  3  0 85   0.20  121 0.25   0.3   3 
ptools3    1  53 C 5.0 2.6  2  10  1  0 89   0.15   52 0.25   0.3   2 

ptools7    2  53 c 2.0 0.4  1   0  1  0 99   0.05    2 0.10   0.3   2 

Host       OS  M Mem InU Lp  Us Sy Wa Id  PhysB Vcsw %istl %bstl %bdon %idon 
------------------------------dedicated-----------------------------------------
ptools6    52   1.1 0.1  1  11  7  0 82   0.50   50   10      5   10   0 
ptools8    52   1.1 0.1  1  11  7  0 82   0.50   60    0      1    -   - 
ptools2    52   1.1 0.1  1  11  7  0 82   0.50  200    0      15  25   10 

Memory pool report

The topasout command generates the Memory pool report that contains information about the memory pools in the CEC and the partitions that belong to the memory pools. The following values are displayed in the header section:
Item Description
Mshr Number of LPARs that are running in the shared-memory mode
Mded Number of LPARs that are running in the dedicated-memory mode
Pools Total number of memory pools in the system
Mpsz Total size of physical memory of all the memory pools in gigabytes
MPuse Total memory used by LPARs associated with all the pools in gigabytes
Entl Total I/O memory entitlement of all of the LPARs in all of the pools in gigabytes
Use Total I/O memory entitlement in use of all of the LPARs in all of the pools in gigabytes
Mon Total monitored memory of the system in gigabytes
InUse Total memory in use of the system in gigabytes
Avl Total free memory available in the system in gigabytes
The following values are displayed in the memory pools section:
Item Description
mpid ID of the memory pool
mpsz Size of the total physical memory of the memory pool in gigabytes
mpus Total memory of the memory pool in use (this value is the sum of the physical memory allocated to all of the LPARs in the pool)
mem Aggregate logical memory size of all of the partitions in the pool in gigabytes
memu Aggregate logical memory used for all of the partitions in the pool in gigabytes
iome Aggregate of I/O memory entitlement that is configured for all of the LPARs in the pool in gigabytes
iomu Aggregate of the I/O memory entitlement that is used for all of the LPARs in the pool in gigabytes
hpi Aggregate number of hypervisor page faults that have occurred for all of the LPARs in the pool
hpit Aggregate amount of time spent waiting for hypervisor page-ins by all of the LPARs in the pool in milliseconds
The following values are displayed in the partitions section:
Item Description
mem Logical memory size of the partition in gigabytes
memu Logical memory that is used for the partition in gigabytes
meml Logical memory that is loaned to the hypervisor by the LPAR
pmem Physical memory allocated to the partition from the memory pool in gigabytes
iom Amount of I/O memory entitlement that is configured for the LPAR in gigabytes
iomu Amount of I/O memory entitlement that is used for the LPAR in gigabytes
hpi Number of hypervisor page faults
hpit Time spent waiting for hypervisor page-ins in milliseconds
vcsw Virtual context switches as an average per second
physb Physical processor busy
%entc Percentage of the processor entitlement that is consumed

Comma-separated reports

The topasout command generates a CEC report that contains data that are separated with comma.

Use the topasout command with the -c flag to generate this report. The output file is written to recordedfilename_01 file.

For example, to generate a report in spreadsheet format for the topas_CEC.070221 file in the /etc/perf/ directory, enter the following command:
topasout -c /etc/perf/topas_CEC.070221

The output file is the topas_CEC.070221_01 file, which locates in the same directory as the original file.

The topas recordings support only the-m mean option.

The following sample shows the output of a topas_CEC report:
#Monitor: topas_CEC recording--- hostname: ptoolsl3 ValueType: mean
Time="2007/03/06 07:19:39", CEC/Lpars/monitored=3.00
Time="2007/03/06 07:19:39", CEC/Lpars/unmonitored=0.00
Time="2007/03/06 07:19:39", CEC/Lpars/shared=1.00
Time="2007/03/06 07:19:39", CEC/Lpars/dedicated=2.00
Time="2007/03/06 07:19:39", ptoolsl1/LPAR/Sys/osver=5.30
Time="2007/03/06 07:19:39", ptoolsl1/LPAR/Sys/shared=1.00
Time="2007/03/06 07:19:39", ptoolsl1/LPAR/Sys/capped=0.00
Time="2007/03/06 07:19:39", ptoolsl1/LPAR/Sys/smt=1.00
… 

Spreadsheet format reports

The topasout command generates a CEC report in spreadsheet format.

Use the topasout command with the -s flag to generate this report. The output file is written to recordedfilename_01 file.

For example, to generate a report in spreadsheet format for the topas_CEC.070221 file in the /etc/perf/ directory, enter the following command:
topasout -s /etc/perf/topas_CEC.070221

The output file is the topas_CEC.070221_01 file, which locates in the same directory as the original file.

The topas recordings can use only the -m mean option.

VIOS report

The VIOS report contains information about Virtual I/O Server/Client throughput. The following column headings are included in a Virtual I/O Server/Client throughput report:
Item Description
Server Name of the VIO Server.
Client Name of the VIO Client.
KBPS Number of kilobytes read and written per second over the monitoring interval. This field is the sum of the values of the KB-R and KB-W metrics.
TPS Number of transfers that are issued per second.
KB-R Number of kilobytes read per second.
KB-W Number of kilobytes written per second.
AQD Average number of requests waiting to be sent.
AQW Average queue that is waiting per request reported in millisecond. The suffix indicates the unit of time. The default time unit is millisecond.
ART Average time to receive a response from the hosting server for the read request sent. The suffix indicates the unit of time. The default time unit is millisecond.
AWT Average time to receive a response from the hosting server for the write request sent. The suffix indicates the unit of time. The default time unit is millisecond.
MRT Maximum time to receive a response from the hosting server for the read request sent. The suffix indicates the unit of time. The default time unit is millisecond.
MWT Maximum time to receive a response from the hosting server for the write request sent. The suffix indicates the unit of time. The default time unit is millisecond.

VIOS adapter report

The VIOS adapter report contains information on virtual I/O server or client (VIOS) adapter and disk details. The following details on the disks are reported:
Item Description
Adapter Name of the server adapter.
Vtargets Name of the virtual target device belonging to the server adapter.
Client_disk Name of the client disk that is mapped to the virtual target device of the server adapter.
The following details of the adapters are displayed:
Item Description
KBPS Amount of data transferred (read or written) in the adapter in kilobytes per second.
TPS Number of transfers per second issued to the adapter.
KB-R Total number of kilobytes read from the adapter.
KB-W Total number of kilobytes written to the adapter.
AQD Average number of requests waiting to be sent to the virtual target device or disk.
AQW Average queue waiting per request reported in millisecond. The suffix indicates the unit of time. The default time unit is millisecond.
ART Average time to receive a response from the hosting server for the read request sent. The suffix indicates the unit of time. The default time unit is millisecond.
AWT Average time to receive a response from the hosting server for the write request sent. The suffix indicates the unit of time. The default time unit is millisecond.
MRT Maximum time to receive a response from the hosting server for the read request sent. The suffix indicates the unit of time. The default time unit is millisecond.
MWT Maximum time to receive a response from the hosting server for the write request sent. The suffix indicates the unit of time. The default time unit is millisecond.
The following details of the virtual target device and the client disk are reported:
Item Description
Busy% Percentage of time that the virtual target device or disk is active.
KBPS Number of kilobytes read and written per second over the monitoring interval. This field is the sum of the value of the KB-R and KB-W metrics.
TPS Number of transfers per second that are issued to the virtual target device or disk. A transfer is an I/O request to the virtual target device or disk. Multiple logical requests can be combined into a single I/O request to the disk. A transfer is of medium size.
KB-R Number of kilobytes read per second from the virtual target device or disk.
KB-W Number of kilobytes written per second to the virtual target device or disk.
AQD Average number of requests waiting to be sent to the virtual target device or disk.
AQW Average queue waiting per request that is reported in milliseconds. The suffix indicates the unit of time. The default time unit is milliseconds.
ART Average time to receive a response from the hosting server for the read request sent. The suffix indicates the unit of time. The default time unit is milliseconds.
AWT Average time to receive a response from the hosting server for the write request sent. The suffix indicates the unit of time. The default time unit is milliseconds.
MRT Maximum time to receive a response from the hosting server for the read request sent. The suffix indicates the unit of time. The default time unit is milliseconds.
MWT Maximum time to receive a response from the hosting server for the write request sent. The suffix indicates the unit of time. The default time unit is milliseconds.

Flags

Item Description
-a The -a flag is used only for nmon analyzer report.
-b time The time in the recorded file that the topasout command begins to generate reports from. The time can either be in the YYMMDDHHMM format or the HHMM format. You must use the same time format for end time if it is specified.

YYMMDD represents year, month, and day. HHMM represents hour and minute.

In HHMM format, the value must range from 0000 through 2359. The default value for begin time is 0000. The report is generated for the first day of the recording within the given time range.

In YYMMDDHHMM format, the default value is the time of the first recorded data in the recording file. The command generates report for the data between the begin and end time range.

-c Specifies that the topasout command should format the output files as comma-separated ASCII. Each line in the output files contains one time stamp and one observation.
-e time The time in the recorded file that the topasout command stop generating reports from. The time can be in the YYMMDDHHMM format or the HHMM format. You must use the same time format for the begin time if it is specified.

YYMMDD represents year, month, and day.

HHMM represents hour and minute.

In YYMMDDHHMM format, the default value is the time of the last recorded data in the recording file. The report is generated for the data between the begin and end date and time range.

In HHMM format, the default value for end time is 2359. The report is generated for the first day of the recording within the given time range.

-i interval The -i flag defines the interval in minute that the topasout command need to average the values. The valid values of the -i flag are 5, 10, 15, 30, or 60. The default value is 5 minutes.
-m type By default, the topasout only outputs the mean values. Other recorded values and the full set for local recordings are available through other options including the min, max, mean, stdev, set, and exp options.
-R type Use the -R flag to specify the type of a report for xmwlm recordings or topasout recordings. The type parameter has the following variables:
summary
Generates Summary report.
detailed
Generates Detailed report.
lan
Generates LAN report.
disk
Generates Disk report.
poolinfo
Generate Shared-processor-pool report.
mempool
Generates memory pool report. If there is no memory pool, the header will be displayed without any values.
adapter
Generates adapter report.
vadapter
Generates virtual adapter report.
vios
Generates Virtual I/O Server/Client throughput report.
vios_adapter
Generates Virtual I/O Server/Client adapter and disk detailed report.

The reports generated with the -R flag are printed to the console.

-s Specifies that topasout should format the output files in a format suitable for input to spreadsheet programs.

Parameters

Item Description
xmwlm_recording_file Specifies that the input file is a recording created using the topasrec/xmwlm command.
topas_recording_file Specifies that the input file is a recording created using the topasrec/topas command.
nmon_recording_file Specifies that the input file is a recording created using the nmon command.

Examples

  1. To generate a Detailed report from an xmwlm recording file from 10:00 a.m. to 11:00 p.m., enter the following command:
    topasout -R detailed  -i 15 –b 1000 -e  2300 /etc/perf/daily/xmwlm.070226
  2. To generate a Summary report from an xmwlm recording file, enter the following command:
    topasout -R summary /etc/perf/daily/xmwlm.070226
  3. To generate a Disk report from an xmwlm recording file, enter the following command:
    topasout -R disk  /etc/perf/daily/xmwlm.070226
  4. To generate a LAN report from an xmwlm recording file, enter the following command:
    topasout -R lan  /etc/perf/daily/xmwlm.070226
  5. To generate an adapter report from an xmwlm recording file, enter the following command:
    topasout -R adapter  /etc/perf/daily/xmwlm.070226
  6. To generate a virtual adapter report from an xmwlm recording file, enter the following command:
    topasout -R vadapter  /etc/perf/daily/xmwlm.070226
  7. To generate a nmon analyzer report from an xmwlm recording file named xmwlm.070226 in the /etc/perf/daily/ directory, enter the following command:
    topasout  –a /etc/perf/daily/xmwlm.070226

    The output is written to /etc/perf/daily/xmwlm.070226.csv

  8. To generate a Shared-processor-pool report from topas CEC recording, enter the following command:
    topasout -R poolinfo /etc/perf/topas_CEC.070302
  9. To generate a Summary report from topas CEC recording from 2:00 p.m. to 4:00 p.m. on the first day of recorded data, enter the following command:
    topasout -R summary  -b 1400 –e 1600 /etc/perf/topas_CEC.070302
  10. To generate a VIOS report from a topas CEC recording, enter the following command:
    topasout -R vios  /etc/perf/topas_CEC.070302
  11. To generate a VIOS adapter report from a topas CEC recording, enter the following command:
    topasout -R vios_adapter  /etc/perf/topas_CEC.070302
  12. To generate a memory pool report from a topas CEC recording, enter the following command:
    topasout -R mempool /etc/perf/topas_CEC.070302
  13. To generate a summary report from a topas CEC recording from 2:00 p.m., March 10, 2008 to 4:00 p.m., March 12,2008, enter the following command:
    topasout -R summary -b 0803101400 -e 0803121600 /etc/perf/ptoolsl1_cec_080310.topas
  14. To generate a detailed report from a topas Cluster recording from 2:00 p.m., March 10, 2008 to 4:00 p.m., March 12,2008, enter the following command:
    topasout -R summary -b 0803101400 -e 0803121600 /etc/perf/ptoolsl1_cluster_080310.topas
  15. To generate a nmon analyzer report from an CEC Recording file named ptoolsl1_cec_080310.topas in the /etc/perf/ directory enter the following command:
    topasout  -a /etc/perf/ptoolsl1_cec_080310.topas
  16. To generate a nmon analyzer report from an Cluster Recording file named ptoolsl1_cluster_080310.topas in the /etc/perf/ directory, enter the following command:
    topasout  -a /etc/perf/ptoolsl1_cluster_080310.topas

Location

/usr/bin/topasout

Files

Item Description
/usr/bin/topas Contains the topas command.
/usr/bin/xmwlm Contains the xmwlm command.
/usr/bin/topasout Contains the topasout command. The topasout command is included in the perfagent.tools fileset.