topas Command
Purpose
Reports selected local and remote system statistics.
Syntax
topas [ -d number_of_monitored_hot_disks ] [ -h ] [ -i monitoring_interval_in_seconds | recording_interval_in_seconds ] [ -n
number_of_monitored_hot_network_interfaces ] [ -p number_of_monitored_hot_processes ] [ -w number_of_monitored_hot_WLM classes ] [ -c
number_of_monitored_hot_CPUs ][ -I remote_polling_interval ] [ -U username_owned_processes ] | [ -Cn | -R [ -o field = value,... ] -D | -L |-P | -W
] [ -m]
Description
The topas command reports selected statistics about the activity on the local system. The command uses the curses library to display its output in a
format suitable for viewing on an 80x25 character-based display or in a window of at least the same size on a graphical display. The topas command
requires the bos.perf.tools and perfagent.tools filesets to be installed on the system.
The topas command can also report a limited set of performance metrics from remote AIX partitions that belong to the same hardware platform. This
support is described in the Cross-Partition View section.
If the topas command is invoked without flags, it runs as if invoked with the following command line:
topas -d20 -i2 -n20 -p20 -w20 -c20
The program extracts statistics from the system with an interval specified by the monitoring_interval_in_seconds argument. The default output, as
shown below, consists of two fixed parts and a variable section. The top two lines at the left of the display show the name of the system the topas
command runs on, the date and time of the last observation, and the monitoring interval.
The second fixed part fills the rightmost 25 positions of the display. It contains five subsections of statistics, as follows:
EVENTS/QUEUES
Displays the per-second frequency of selected system-global events and the average size of the thread run and wait queues:
Cswitch
The number of context switches per second over the monitoring interval.
Syscalls
The total number of system calls per second executed over the monitoring interval.
Reads
The number of read system calls per second executed over the monitoring interval.
Writes
The number of write system calls per second executed over the monitoring interval.
Forks
The number of fork system calls per second executed over the monitoring interval.
Execs
The number of exec system calls per second executed over the monitoring interval.
Runqueue
The average number of threads that were ready to run but were waiting for a processor to become available.
Waitqueue
The average number of threads that were waiting for paging to complete.
FILE/TTY
Displays the per-second frequency of selected file and tty statistics.
Readch
The number of bytes read per second through the read system call over the monitoring interval.
Writech
The number of bytes written per second through the write system call over the monitoring interval.
Rawin
The number of raw bytes read per second from TTYs over the monitoring interval.
Ttyout
The number of bytes written to TTYs per second over the monitoring interval.
Igets
The number of calls per second to the inode lookup routines over the monitoring interval.
Namei
The number of calls per second to the pathname lookup routines over the monitoring interval.
Dirblk
The number of directory blocks scanned per second by the directory search routine over the monitoring interval.
PAGING
Displays the per-second frequency of paging statistics.
Faults
Total number of page faults taken per second over the monitoring interval. This includes page faults that do not cause paging activity.
Steals
Physical memory 4K frames stolen per second by the virtual memory manager over the monitoring interval.
PgspIn
Number of 4K pages read from paging space per second over the monitoring interval.
PgspOut
Number of 4K pages written to paging space per second over the monitoring interval.
PageIn
Number of 4 K pages read per second over the monitoring interval. This includes paging activity associated with reading from file
systems. Subtract PgspIn from this value to get the number of 4K pages read from file systems per second over the monitoring interval.
PageOut
Number of 4K pages written per second over the monitoring interval. This includes paging activity associated with writing to file
systems. Subtract PgspOut from this value to get the number of 4K pages written to file systems per second over the monitoring interval.
Sios
The number of I/O requests per second issued by the virtual memory manager over the monitoring interval.
MEMORY
Displays the real memory size and the distribution of memory in use.
Real,MB
The size of real memory in megabytes.
% Comp
The percentage of real memory currently allocated to computational page frames. Computational page frames are generally those that are
backed by paging space.
% Noncomp
The percentage of real memory currently allocated to non-computational frames. Non-computational page frames are generally those that
are backed by file space, either data files, executable files, or shared library files.
% Client
The percentage of real memory currently allocated to cache remotely mounted files.
PAGING SPACE
Display size and use of paging space.
Size,MB
The sum of all paging spaces on the system, in megabytes.
% Used
The percentage of total paging space currently in use.
% Free
The percentage of total paging space currently free.
NFS
Display NFS stats in calls per second
* Server V2 calls/sec
* Client V2 calls/sec
* Server V3 calls/sec
* Client V3 calls/sec
The variable part of the topas display can have one, two, three, four, or five subsections. If more than one subsection displays, they are always
shown in the following order:
* Processor
* Network Interfaces
* Physical Disks
* Workload Manager Classes
* Processes
When the topas command is started, it displays all subsections for which hot entities are monitored. The exception to this is the Workload Manager
(WLM) Classes subsection, which is displayed only when WLM is active.
CPU Utilization
This subsection displays a bar chart showing cumulative processor usage. If more than one processor exists, a list of processors can be
displayed by pressing the c key twice. Pressing the c key only once will turn this subsection off. The following fields are displayed by both
formats:
User%
This shows the percent of processor used by programs executing in user mode. (Default sorted by User%)
Kern%
This shows the percent of processor used by programs executing in kernel mode.
Wait%
This shows the percent of time spent waiting for IO.
Idle%
This shows the percent of time the processor(s) is idle.
Physc
Number of physical processors consumed. Displayed only if the partition is running with shared processor.
%Entc
Percentage of entitled capacity consumed. Displayed only if the partition is running with shared processor.
When this subsection first displays the list of hot processors, the list is sorted by the User% field. However, the list can be sorted by the
other fields by moving the cursor to the top of the desired column.
Network Interfaces
This subsection displays a list of hot network interfaces. The maximum number of interfaces displayed is the number of hot interfaces being
monitored, as specified with the -n flag. A smaller number of interfaces will be displayed if other subsections are also being displayed.
Pressing the n key turns off this subsection. Pressing the n key again shows a one-line report summary of the activity for all network
interfaces. Both reports display the following fields:
Interf
The name of the network interface.
KBPS
The total throughput in megabytes per second over the monitoring interval. This field is the sum of kilobytes received and kilobytes
sent per second.
I-Pack
The number of data packets received per second over the monitoring interval.
O-Pack
The number of data packets sent per second over the monitoring interval.
KB-In
The number of kilobytes received per second over the monitoring interval.
KB-Out
The number of kilobytes sent per second over the monitoring interval.
When this subsection first displays the list of hot network interfaces, the list is sorted by the KBPS field. However, the list can be sorted
by the other fields by moving the cursor to the top of the desired column. Sorting is only valid for up to 16 network adapters.
Physical Disks
This subsection displays a list of hot physical disks. The maximum number of physical disks displayed is the number of hot physical disks being
monitored as specified with the -d flag. A smaller number of physical disks will be displayed if other subsections are also being displayed.
Pressing the d key turns off this subsection. Pressing the d key again shows a one-line report summary of the activity for all physical disks.
Both reports display the following fields:
Disk
The name of the physical disk.
Busy%
Indicates the percentage of time the physical disk was active (bandwidth use for the drive).
KBPS
The number of kilobytes read and written per second over the monitoring interval. This field is the sum of KB-Read and KB-Read.
TPS
The number of transfers per second that were 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. A transfer is of indeterminate size.
KB-Read
The number of kilobytes read per second from the physical disk.
K -Writ
The number of kilobytes written per second to the physical disk.
When this subsection first displays the list of hot physical disks, the list is sorted by the KBPS field. However, the list can be sorted by
the other fields by moving the cursor to the top of the desired column. Sorting is only valid for up to 128 physical disks.
WLM Classes
This subsection displays a list of hot Workload Manager (WLM) Classes. The maximum number of WLM classes displayed is the number of hot WLM
classes being monitored as specified with the -w flag. A smaller number of classes will be displayed if other subsections are also being
displayed. Pressing the w key turns off this subsection. The following fields are displayed for each class:
% CPU Utilization
The average processor use of the WLM class over the monitoring interval.
% Mem Utilization
The average memory use of the WLM class over the monitoring interval.
% Blk I/O
The average percent of block I/O of the WLM class over the monitoring interval.
When this subsection first displays the list of hot WLM classes, the list will be sorted by the CPU% field. However, the list can be sorted by
the other fields by moving the cursor to the top of the desired column.
Processes
This subsection displays a list of hot processes. The maximum number of processes displayed is the number of hot processes being monitored as
specified with the -p flag. A smaller number of processes will be displayed if other subsections are also being displayed. Pressing the p key
turns off this subsection. The processes are sorted by their processor usage over the monitoring interval. The following fields are displayed
for each process:
Name
The name of the executable program executing in the process. The name is stripped of any pathname and argument information and truncated
to 9 characters in length.
Process ID
The process ID of the process.
% CPU Utilization
The average processor use of the process over the monitoring interval. The first time a process is shown, this value is the average
processor use over the lifetime of the process.
Paging Space Used
The size of the paging space allocated to this process. This can be considered an expression of the footprint of the process but does
not include the memory used to keep the executable program and any shared libraries it may depend on.
Process Owner (if the WLM section is off)
The user name of the user who owns the process.
Workload Manager (WLM) Class (if the WLM section is on)
The WLM class to which the process belongs.
Cross-Partition View and Recording
This panel displays metrics similar to the lparstat command for all the AIX partitions it can identify as belonging to the same hardware platform.
Dedicated and shared partitions are displayed in separate sections with appropriate metrics. The top section represents aggregated data from the
partition set to show overall partition, memory, and processor activity.
Remote enablement for this panel to collect from other partitions requires that latest updates to the perfagent.tools and bos.perf.tools used to
support this function. For earlier versions of AIX, the topas command also collects remote data from partitions that have the Performance Aide
product (perfagent.server) installed. topas -C may not be able to locate partitions residing on other subnets. To circumvent this, create a
$HOME/Rsi.hosts file containing the fully qualified host names for each partition (including domains), one host per line.
Besides the display option, the topas -R option runs the topas command in the background (with no display) and record the Cross-Partition data.
Recordings are made to the /etc/perf/ directory, and are of the form topas_cec.YYMMDD. Recordings cover single-day periods (24 hours) and are
retained for 8 days before they are automatically deleted. This allows a week's worth of data to be retained on the system at all times. The
topasout command is used to convert these recordings into various text-based reports.
If you use the topas -R function on an LPAR-capable system, you should identify a single partition for collecting and recording this data. The
/usr/lpp/perfagent/config_topas.sh script can be used to automatically update the system /etc/inittab file to run topas -R on startup.
The following metrics display in the initial Cross-Partition panel. Additional metrics with full descriptive labels can be displayed by using the
key toggles identified in the Additional Cross-Partition Panel Subcommands section:
Partition totals:
Shr
Number of shared partitions
Ded
Number of dedicated partitions
Memory (in GB)
Mon
Monitored partitions total memory
Avl
Memory available to partition set
InUse
Memory in use on monitored partitions
Processor:
Shr
Number of shared processors
Ded
Number of dedicated processors
PSz
Active physical processors in the physical shared processor pool being used by this LPAR
APP
Available physical processors in the shared pool. This is equal to the idle cycles pool reported as a number of processors
Don
Total number of processors donated to the pool
Shr_PhysB
Total number of physical processors that are busy for all shared partitions
Ded_PhysB
Total number of physical processors that are busy for all dedicated partitions
Individual partition data:
Host
Hostname
OS
Operating system level
M
Mode
For shared partitions:
* C - SMT enabled and capped
* c - SMT disabled and capped
* U - SMT enabled and uncapped
* u - SMT disabled and uncapped
For dedicated partitions:
* S - SMT enabled
* d - SMT disabled and donating
* D - SMT enabled and donating
* ' ' (blank) - SMT disabled and not donating
Mem
Total memory in GB.
InU
Memory in use in GB.
Lp
Number of logical processors
Us
Percentage of processor used by programs executing in user mode.
Sy
Percentage of processor used by programs executing in kernel mode.
Wa
Percentage of time spent waiting for I/O.
Id
Percentage of time the processor(s) is idle.
PhysB
Physical Processors Busy
Ent
Entitlement granted (shared-only)
%Entc
Percent 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 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. Requirement: At least one partition to be monitored must have
Pool Utilization Authority (PUA) configured for pool information metrics to be collected.
For cross-partition monitoring/recording, some global data is not available from any partition. The -o options allow you to specify these fields in
the command line. Optionally, you can configure a system to allow the topas command to query the HMC directly for this information. This requires
the following steps:
1 Install OpenSSH at the partition.
2 Enable remote command support on the HMC for user hscroot to allow ssh connections to be opened from the partition.
3 Configure ssh on the HMC to not require a password for the HMC user hscroot when queried from the selected partition. This requires the
.ssh/authorized_keys2 on the HMC for user login hscroot.
4 Run ssh -l hscroot hmc_address date from the partition to confirm whether the date is displayed without requiring that a password be entered.
5 Utilize the topas -o options described in the usage table to specify the managed system and HMC names when running the topas command.
Restriction: This functionality is currently available only for HMC version 5 and above, and should only be enabled after careful consideration
of any security implications.
The following displays when press the g key in the initial screen, which brings the cross partition view with detailed headers:
Topas CEC Monitor Interval: 10 Mon Jan 22 00:08:00 2007
Partition Info Memory (GB) Processor Virtual Pools : 2
Monitored : 2 Monitored : 6.2 Monitored :2.0 Avail Pool Proc: 5
UnMonitored: - UnMonitored: - UnMonitored: - Shr Physical Busy: 0.00
Shared : 0 Available : - Available : - Ded Physical Busy: 0.05
Uncapped : 0 UnAllocated: - UnAllocated: - Donated Phys. CPUs: 0.00
Capped : 2 Consumed : 1.9 Shared : 0 Stolen Phys. CPUs : 0.01
Dedicated : 2 Dedicated : 2 Hypervisor
Donating : 0 Donated : 0 Virt. Context Switch: 347
Pool Size : 0 Phantom Interrupts : 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 1 2 0 96 0.01 398 0.20 5.3 0k
Host OS M Mem InU Lp Us Sy Wa Id PhysB Vcsw %istl %bstl %bdon %idon
------------------------------------dedicated-----------------------------------
ptools1 A54 S 3.1 0.9 2 0 0 0 99 0.00 177 0.1 0.0 0.0 0.0
ptoolsl3 A54 S 3.1 0.9 2 0 0 0 99 0.00 170 0.2 0.0 0.0 0.0
The following headers are in the previous screen:
Partition Info:
Monitored
Number of partitions monitored
Unmonitored
Number of partitions unmonitored
Shared
Number of shared partitions
Uncapped
Number of uncapped shared partitions
Capped
Number of capped partitions
Dedicated
Number of dedicated partitions
Donating
Number of partitions that are currently donating
Memory:
Monitored
Total monitored memory
UnMonitored
Total unmonitored memory
Available
Total available memory
UnAllocated
Total memory not allocated to any partition
Consumed
Total memory consumed by the partitions
Processor:
Monitored
Number of physical processors monitored
UnMonitored
Number of physical processors not monitored
Available
Number of physical processors available in CEC system
UnAllocated
Number of physical processors 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 the partitions donating
Pool Size
Number of active shared processors in all the pools
Avail Proc Pool
Available physical processors in pool
Shr Physical Busy
Sum of physical busy of all shared partitions
Ded Physical Busy
Sum of dedicated busy of all dedicated partitions
Donated Phys. processors
Sum of donated processor cycles from all partitions reported as a number of processors
Stolen Phys. processors
Sum of stolen processor cycles from all partitions reported as a number of processors
Virtual Pools
Number of virtual pools
Virt. Context Switch
Total number of virtual context switches per second in the monitoring interval
Phantom Interrupts
Total number of phantom interrupts per second in the monitoring interval
When the topas command is running inside any cross partition view, press the p key to bring up the pool panel. The following example shows the
display:
pool psize entc maxc physb app mem muse
0 3.0 2.0 4.0 0.1 2.0 1.0 1.5
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
User can scroll up or down in the pool ID column and press the f key to list only the shared partitions belonging to the poolid where cursor is
positioned. The following example shows the display:
psize
Effective maximum capacity of the pool
entc
Entitled capacity of the pool
maxc
Maximum capacity of the pool
physb
Summation of physical busy of processors in shared partitions of a pool
app
Available physical processor in the pool
mem
Sum of monitored memory for all shared partitions in the pool
muse
Sum of memory consumed for all shared partitions in the pool
Implementation Specifics
Disks and network adapters added after starting topas or any other SPMI consumer will not be reflected in topas. You must stop topas and all clients
that use SPMI and then restart after the changes to disks and network adapters are made.
Sample Default Output
The following is example shows the display generated by the topas command:
Topas Monitor for host: niller EVENTS/QUEUES FILE/TTY
Mon Mar 13 15:56:32 2000 Interval: 2 Cswitch 113 Readch 1853576
Syscall 2510 Writech 49883
CPU User% Kern% Wait% Idle% Reads 466 Rawin 0
cpu0 7.0 4.0 0.0 89.0 Writes 12 Ttyout 706
cpu1 1.0 8.0 0.0 91.0 Forks 0 Igets 0
cpu2 0.0 0.0 0.0 100.0 Execs 0 Namei 0
Runqueue 0.0 Dirblk 0
Waitqueue 0.0
Interf KBPS I-Pack O-Pack KB-In KB-Out
lo0 100.4 45.7 45.7 50.2 50.2 PAGING MEMORY
tr0 2.0 4.4 3.4 1.4 0.6 Faults 1 Real,MB 255
Steals 0 % Comp 81.0
Disk Busy% KBPS TPS KB-Read KB-Writ PgspIn 0 % Noncomp 19.0
hdisk0 0.0 0.0 0.0 0.0 0.0 PgspOut 0 % Client 3.0
hdisk1 0.0 0.0 0.0 0.0 0.0 PageIn 0
PageOut 0 PAGING SPACE
WLM-Class (Active) CPU% Mem% Disk% Sios 0 Size,MB 0
System 8 41 12 % Used
Shared 1 24 9 % Free
NFS calls/sec
Name PID CPU% PgSP Class ServerV2 0 Press:
topas (35242) 3.0 0.3 System ClientV2 0 "h" for help.
X (3622) 1.4 44.4 System ServerV3 0 "q" to quit.
notes (25306) 1.3 123.3 System ClientV3 0
Sample Full Screen Process Output
The following is example shows the full screen process output generated by the topas command:
Topas Monitor for host: mothra Interval: 2 Wed Nov 8 12:27:34 2000
DATA TEXT PAGE PGFAULTS
USER PID PPID PRI NI RES RES SPACE TIME CPU% I/O OTH COMMAND
root 1806 0 37 41 16 3374 16 13:25 1.0 0 0 gil
root 1032 0 16 41 3 3374 3 0:00 0.0 0 0 lrud
root 1290 0 60 41 4 3374 4 0:02 0.0 0 0 xmgc
root 1548 0 36 41 4 3374 4 0:26 0.0 0 0 netm
root 1 0 60 20 197 9 180 0:24 0.0 0 0 init
root 2064 0 16 41 4 3374 4 0:04 0.0 0 0 wlmsched
root 2698 1 60 20 14 2 14 0:00 0.0 0 0 shlap
root 3144 1 60 20 40 1 36 5:19 0.0 0 0 syncd
root 3362 0 60 20 4 3374 4 0:00 0.0 0 0 lvmbb
root 3666 1 60 20 135 23 123 0:00 0.0 0 0 errdemon
root 3982 0 60 20 4 3374 4 0:01 0.0 0 0 rtcmd
root 4644 1 17 20 6 3374 6 0:00 0.0 0 0 dog
root 4912 1 60 20 106 13 85 0:00 0.0 0 0 srcmstr
root 5202 4912 60 20 94 8 84 0:01 0.0 0 0 syslogd
root 5426 4912 60 20 195 76 181 0:12 0.0 0 0 sendmail
root 5678 4912 60 20 161 11 147 0:01 0.0 0 0 portmap
root 5934 4912 60 20 103 11 88 0:00 0.0 0 0 inetd
root 6192 4912 60 20 217 61 188 0:21 0.0 0 0 snmpd
root 6450 4912 60 20 137 10 116 0:00 0.0 0 0 dpid2
root 6708 4912 60 20 157 29 139 0:06 0.0 0 0 hostmibd
root 0 0 16 41 3 3374 3 7:08 0.0 0 0
root 6990 1 60 20 106 10 86 0:06 0.0 0 0 cron
Sample Full-Screen Workload Manager Classes Output
The following is example shows the full-screen Workload Manager classes output generated by the topas command:
Topas Monitor for host: mothra Interval: 2 Wed Nov 8 12:30:54 2000
WLM-Class (Active) CPU% Mem% Disk-I/O%
System 0 0 0
Shared 0 0 0
Default 0 0 0
Unmanaged 0 0 0
Unclassified 0 0 0
==============================================================================
DATA TEXT PAGE PGFAULTS
USER PID PPID PRI NI RES RES SPACE TIME CPU% I/O OTH COMMAND
root 1 0 108 20 197 9 180 0:24 0.0 0 0 init
root 1032 0 16 41 3 3374 3 0:00 0.0 0 0 lrud
root 1290 0 60 41 4 3374 4 0:02 0.0 0 0 xmgc
root 1548 0 36 41 4 3374 4 0:26 0.0 0 0 netm
root 1806 0 37 41 16 3374 16 13:25 0.0 0 0 gil
root 2064 0 16 41 4 3374 4 0:04 0.0 0 0 wlmsched
root 2698 1 108 20 14 2 14 0:00 0.0 0 0 shlap
root 3144 1 108 20 40 1 36 5:19 0.0 0 0 syncd
root 3362 0 108 20 4 3374 4 0:00 0.0 0 0 lvmbb
root 3666 1 108 20 135 23 123 0:00 0.0 0 0 errdemon
root 3982 0 108 20 4 3374 4 0:01 0.0 0 0 rtcmd
Sample Full-Screen Cross-Partition Output
The following is example shows the full-screen cross-partition output generated by the topas command:
Topas CEC Monitor Interval: 10 Wed Mar 6 14:30:10 2005
Partitions Memory (GB) Processors
Shr: 4 Mon: 24 InUse: 14 Mon: 8 PSz: 4 Shr_PhysB: 1.7
Ded: 4 Avl: 24 Avl: 8 APP: 4 Ded_PhysB: 4.1
Host OS M Mem InU Lp Us Sy Wa Id PhysB Ent %EntC Vcsw PhI
--------------------------------shared------------------------------------------
ptools1 A53 u 1.1 0.4 4 15 3 0 82 1.30 0.50 22.0 200 5
ptools5 A53 U 12 10 1 12 3 0 85 0.20 0.25 0.3 121 3
ptools3 A53 C 5.0 2.6 1 10 1 0 89 0.15 0.25 0.3 52 2
ptools7 A53 c 2.0 0.4 1 0 1 0 99 0.05 0.10 0.3 112 2
-------------------------------dedicated----------------------------------------
ptools4 A53 S 0.6 0.3 2 12 3 0 85 0.60
ptools6 A52 1.1 0.1 1 11 7 0 82 0.50
ptools8 A52 1.1 0.1 1 11 7 0 82 0.50
ptools2 A52 1.1 0.1 1 11 7 0 82 0.50
Flags
-c
Specifies the number of hot processors to be monitored. This is also the maximum number of processors displayed when enough room is available
on the screen. If this number exceeds the number of processors available, only the installed processors will be monitored and displayed. If
this argument is omitted, a default of 2 is assumed. If a value of 0 (zero) is specified, no processor information is monitored.
-C
Displays the Cross-Partition panel. The topas command collects a set of metrics from AIX partitions running on the same hardware platform. The
metrics are similar to those collected by the lparstat command. Dedicated and shared partitions are displayed, and a set of aggregated values
provide an overview of the entire hardware systems partition set. Certain values only available from the HMC platform may be set via the line
command if an HMC connection is not available:
* -o availmem = [total amount of memory allocated to all partitions, in GB]
* -o unavailmem = [total amount of memory unallocated from the HMC, in GB]
* -o availcpu = [total number of physical processors allocated for all partitions]
* -o unavailcpu = [total number of physical processors unallocated from the HMC]
* -o poolsize = [defined Pool Size required if HMC Processor Utilization Authority restricts access]
* -o partitions = [number of partitions defined on the HMC]
* -o reconfig = [number of seconds between checking for HMC configuration changes. Allowed values are 30, 60, 90, 120, 180, 240, 300
seconds. The default is 60 seconds]
-D
Displays the Disk Metrics display. The display reports disk service times, disk queuing metrics, and disk throughput. The following metrics are
reported:
AQD
Average number of requests waiting to be sent to disk.
AQW
Average queue wait per request reported in milliseconds.
ART
Indicates the 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
Indicates the 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
Indicates the 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
Indicates the 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.
-d
Specifies the number of disks to be monitored. This is also the maximum number of disks displayed when enough room is available on the screen.
When this number exceeds the number of disks installed, only the installed disks will be monitored and displayed. If this argument is omitted,
a default of 2 is assumed. If a value of 0 (zero) is specified, no disk information is monitored.
-h
Displays help information in the following format:
usage: topas [-d number-of-monitored-hot-disks]
[-h]
[-i monitoring-interval_in_seconds]
[-n number-of-monitored-hot-network-interfaces]
[-p number-of-monitored-hot-processes]
[-w number-of-monitored-hot-WLM classes]
[-c number-of-monitored-hot-CPUs]
[-U username_owned_processes]
[-D|-P|-W|-L]
[-m]
-i
Sets the monitoring interval in seconds. The default is 2 seconds. If used with the -R mode, this becomes the recording interval for partition
metrics. The default is 300 seconds. Valid values are 10, 15, 30, 60, 120, and 300 seconds.
-I remote_polling_interval
For cross-partition display, sets the sampling interval to collect data from remote partitions. The default is 10 seconds. Values of 10, 15,
30, 60 and 120 seconds are allowed.
-L
Displays the logical partition display. This display reports similar data to what is provided to mpstat and lparstat.
-m
Displays in monochrome mode (no colors).
-n
Specifies the number of hot network interfaces to be monitored. This is also the maximum number of network interfaces displayed when enough
room is available on the screen. When this number exceeds the number of network interfaces installed, only the installed network interfaces
will be monitored and displayed. If this argument is omitted, a default of 2 is assumed. If a value of 0 (zero) is specified, no network
information is monitored.
-P
Similar to the ps command, the -P flag displays the full-screen process display. This display shows a list of the busiest processes, similar to
the process subsection on the default display, only with more columns showing more metrics per process. This list can be sorted by any column.
Following are the metrics displayed.
USER
The login name of the process owner.
PID
The process ID of the process.
PPID
The process ID of the parent process.
PRI
The priority of the process or kernel thread; higher numbers mean lower priority.
NI
The process priority specified with the nice command; used in calculating priority for the sched other policy.
Data RSS
The real-memory data (resident set) size of the process (4KB pages).
TEXT RSS
The real-memory text (resident set) size of the process (4KB pages).
PAGE SPACE
Virtual working set size used by process (4KB pages). Note: The true paging space allocations per process are not available using the
topas command. Refer to the svmon command for more detailed reports.
TIME
The total execution time for the process.
CPU%
The percentage of processor usage.
PGFAULTS
Shows the number of I/O and other page faults.
COMMAND
Contains the command name. Truncates the command name to 9 characters.
-p
Specifies the number of hot processes to be monitored. This is also the maximum number of processes shown when enough room is available on the
screen. If this argument is omitted, a default of 20 is assumed. If a value of 0 is specified, no process information will be monitored.
Retrieval of process information constitutes the majority of the topas overhead. If process information is not required, always use this option
to specify that you do not want process information.
-R
Records the Cross-Partition data. The topas command collects a set of metrics from AIX partitions running on the same hardware platform. The
metrics are similar to those collected by the lparstat command. Dedicated and shared partition data is recorded, and a set of aggregated values
provide an overview of the entire hardware systems partition set.
Recordings cover only a 24-hour period and are retained for a period of 2 to 7 days. Recordings are made to the /etc/perf/ directory and are of
the form topas_cec.YYMMDD.
If you have configured the topas hosting partition to query the HMC (see above), the following options are used to specify the managed system
and HMC hostname:
* -o managedsys = [managed system name under which this partition is configured, available via: lsrsrc -dx IBM.ManagementServer Hostname]
* -o hmc = [HMC name under which this partition is configured]
The HMC name as a parameter is optional and can be used when the resource class IBM.ManagementServer is not configured in RMC. It is
recommended that the IP address of the HMC be provided instead of the host name. The managed system name needs to be specified as a parameter
for the topas command for the query.
If the HMC platform cannot be configured for automatic queries, then the global data fields not available to the local partition may be set
using the following options:
* -o availmem = [total amount of memory allocated to all partitions, in GB]
* -o unavailmem = [total amount of memory unallocated from the HMC, in GB]
* -o availcpu = [total number of physical processors allocated for all partitions]
* -o unavailcpu = [total number of physical processors unallocated from the HMC]
* -o poolsize = [defined Pool Size required if HMC Processor Utilization Authority restricts access]
* -o partitions = [number of partitions defined on the HMC]
* -o reconfig = [number of seconds between checking for HMC configuration changes. Valid values are 30, 60, 90, 120, 180, 240, and 300
seconds. The default is 300 seconds]
* -o dir = [optional recording directory path]
-U username
Shows username owned processes with the -P flag. Only processes owned by username will be shown in the All Process Display.
-W
Displays the full-screen WLM class display, which is a split display. The top part of the display shows a list of hot WLM classes, similar to
the WLM classes subsection on the default display, but with enough space available to display the full class names. This list can be sorted on
any column.
The bottom part of the display shows a list of busiest processes, similar to the full screen process display, but only displays processes
belonging to one WLM class (selected with the f key).
-w
Specifies the number of hot Workload Manager (WLM) classes to be monitored. This is also the maximum number of WLM classes displayed when
enough room is available on the screen. If this number exceeds the number of WLM classes installed, only the installed WLM classes will be
monitored and displayed. If this argument is omitted, a default of 2 is assumed. If a value of 0 (zero) is specified, no WLM class information
is monitored.
General Subcommands
While topas is running, it accepts one-character subcommands. Each time the monitoring interval elapses, the program checks for one of the following
subcommands and responds to the action requested.
a
The a key shows all of the variable subsections being monitored (processor, network, disk, WLM, and process). Pressing the a key always returns
the topas command to the initial main display.
c
The c key toggles the processor subsection between the cumulative report, off, and a list of the busiest processors. The number of busiest
processors displayed will depend upon the space available on the screen.
C
The uppercase C key activates the Cross-Partition panel, or if currently active, resets the panel to display the global summary, dedicated, and
shared sections. See the Additional Cross-Partition Panel Subcommands section below for options specific to this panel.
d
The d key toggles the disk subsection between a list of busiest disks, off, and the report on the total disk activity of the system. The number
of busiest disks displayed will depend upon the space available on the screen.
D
The uppercase D key replaces the current display with the Disk Metric display. This display offers additional information about disk access
times and disk queuing. When the D key is pressed again, it toggles back to the default main screen.
f
Moving the cursor over a WLM class and pressing the f key displays the list of top processes in the class at the bottom of the WLM screen. This
key is valid only when topas is in the full-screen WLM display (by using the W key or the -W flag).
h
Show the help screen.
H
Shows the help screen for the local panel, if available.
L
The uppercase L key replaces the current display with the logical partition display; LPAR, Micro-Partitioning, and simultaneous multi-threading
metrics similar to what lparstat and mpstat provide are displayed.
n
The n key toggles the network interfaces subsection between a list of busiest interfaces, off, and the report on the total network activity of
the system. The number of busiest interfaces displayed will depend upon the space available on the screen.
p
The p key toggles the hot processes subsection on and off. The number of busiest processes displayed will depend upon the space available on
the screen.
P
The uppercase P key replaces the default display with the full-screen process display. This display provides more detailed information about
processes running on the system than the process section of the main display. When the P key is pressed again, it toggles back to the default
main display.
q
Quit the program.
r
Refresh the display.
w
The w key toggles the Workload Manager (WLM) classes subsection on and off. The number of busiest WLM classes displayed will depend upon the
space available on the screen.
W
The uppercase W key replaces the default display with the full-screen WLM class display. This display provides more detailed information about
WLM classes and processes assigned to classes. When the W key is pressed again, it toggles back to the default main display.
Arrow and Tab keys
Subsections from the main display such as the processor, Network, Disk, WLM Classes, and the full-screen WLM and Process displays can be sorted
by different criteria. Positioning the cursor over a column activates sorting on that column. The entries are always sorted from highest to
lowest value. The cursor can be moved by using the Tab key or the arrow keys. Sorting is only valid for 128 disks and 16 network adapters.
Additional Cross-Partition Panel Subcommands
When the topas Cross-Partition panel is active, it will accept the following additional one-character subcommands. Each time the monitoring interval
elapses, the program checks for one of the following subcommands and responds to any requested action.
d
The d key toggles the dedicated partition section on and off.
g
The g key toggles the top global section of the panel between brief listing, detailed listing, and off.
r
The r key forces topas to search the for HMC configuration changes if a connection is available. This includes the discovery of new partitions,
processors, or memory allocations.
s
The s key toggles the shared partition section on and off.
p
The p key toggles the pool panel section on or off. Inside the pool panel, user can select one pool ID and press the f key to list the shared
partitions belonging to the pool.
Examples
1 To display up to twenty "hot" disks every five seconds and omit network interface, WLM classes, and process information, type:
topas -i5 -n0 -p0 -w0
2 To display the five most active processes and up to twenty most active WLM classes (which is the default when omitting the -w flag) but no
network or disk information, type:
topas -p5 -n0 -d0
3 To run the program with default options, type:
topas
4 To go directly to the process display, type:
topas -P
5 To go directly to the WLM classes display, type:
topas -W
6 To go directly to the logical partition display, type:
topas -L
7 To go directly to the disk metric display, type:
topas -D
Files
/usr/bin/topas
Contains the topas command.
Related Information
The iostat command, vmstat command, wlmstat command, sar command, ps command, mpstat command, lparstat command.
System Performance Measurement Interface in the Performance Toolbox Version 2 and 3 for AIX: Guide and Reference
The lchmon sample program that ships with perfagent.server.
Refer to the /usr/lpp/perfagent/README.perfagent.tools file for additional information on topas.
Refer to the /usr/lpp/perfagent/config_topas.sh script for configuring topas recording as an /etc/inittab process.