Shulou(Shulou.com)06/02 Report--

Topas command

Use

Reports statistics for the selected local and remote systems.

Grammar

Topas [- d hotdisk] [- f hotfs] [- h] [- I interval] [- n hotni] [- p hotprocess] [- w hotwlmclass] [- c hotprocessor] [- Iremotepollinterval] [- @ [wparname]] [- U username] | [- C-D |-G |-F |-L |-P |-V |-T |-M |-E |-W] [- m]

Restrictions: you cannot use the-C,-L,-E,-V,-v,-T,-t,-w,-W,-I, and-@ options when issuing commands from workload partitions.

Description

The topas command reports statistics for selected local system activity. This command uses the curses library to display its output in an appropriate format suitable for viewing on an 80x25 character-based display or in a window of at least the same size graphics 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 remote AIX from the same hardware platform. A limited collection of performance metrics on a partition. This support is described in the cross-partition view and cluster utilization view section.

Note: the tool must be restarted to reflect any new dynamic configuration changes made to the system.

The topas-D command reports disk details. This report is described in the disk panel section. You can run subcommands from the disks panel to display the following views:

Adapter panel

Specify by pressing d. This panel provides more information about the adapter and the disks that belong to the selected adapter.

Virtual adapter panel

Specify by pressing d and then v. This panel provides details about disk-related virtual adapters.

MPIO panel

Specify by pressing the m key. This panel provides details of the disk and path.

Panel freeze

Specify by pressing the Spacebar on the keyboard. The spacebar acts as a switch to freeze the topas panel.

Rolling

The page up and page down keys are used to scroll through the data.

Restrictions: adapter panels, virtual adapter panels, and MPIO panels are restricted to WPAR.

If you invoke the topas command without a flag, it will run in the same way as the following command:

Topas-d20-i2-n20-p20-w20-c20-f0

Note: when the migration or hibernation of the partition is complete, the central electronic complex (CEC) or cluster panel will be rederived. During a migration or hibernation event, all other behaviors of CEC and any other panels remain the same.

The program extracts statistics from the system at intervals specified by the monitoring_interval_in_seconds parameter. The default output is shown below, which contains two fixed parts and one variable part. The top two lines on the left side of the monitor show the name of the system on which the topas command was run, the date and time it was last viewed, and the monitoring interval.

The second fixed part occupies 25 positions at the rightmost end of the monitor. It contains the following statistics subsections:

Project

Description

Event / queue

Displays the frequency per second of selected system global events, the average size of thread runs, and wait queues:

Cswitch

The number of contexts changed per second during the monitoring interval.

Syscalls

The total number of system calls running per second during the monitoring interval.

Reads

The number of read system calls running per second during the monitoring interval.

Writes

The number of write system calls running per second during the monitoring interval.

Forks

The number of derived system calls running per second during the monitoring interval.

Execs

The number of execution system calls running per second during the monitoring interval.

Runqueue

The average number of threads ready to run but need to wait for the processor to be available.

Waitqueue

The average number of threads waiting for page scheduling to complete.

File / TTY

Displays the frequency per second of the selected file and TTY statistics. The following data will be reported:

Readch

Read the number of bytes per second read by the system call during the monitoring interval.

Writech

The number of bytes written by the system call per second during the monitoring interval.

Rawin

The original number of bytes read from the TTY per second during the monitoring interval.

Ttyout

The number of bytes written to the TTY per second during the monitoring interval.

Igets

The number of times the Inode lookup routine was called per second during the monitoring interval.

Namei

The number of times the pathname lookup routine was called per second during the monitoring interval.

Dirblk

The number of directory blocks scanned per second by the directory search routine during the monitoring interval.

Page scheduling

Displays the frequency per second of page scheduling statistics. The following data will be reported:

Faults

The total number of page faults per second during the monitoring interval. This includes a page fault that cannot activate page scheduling.

Steals

During the monitoring interval, 4K frames of physical memory per second are occupied by the virtual memory manager.

PgspIn

The number of 4K pages per second read from the paging space during the monitoring interval.

PgspOut

The number of 4K pages written to the paging space per second during the monitoring interval.

PageIn

The number of 4K pages read per second during the monitoring interval. This includes page scheduling activities related to reading from the file system. Subtract PgspIn from this value to get the number of 4K pages read from the file system per second during the monitoring interval.

PageOut

The number of 4K pages written per second during the monitoring interval. This includes page scheduling activities related to writing to the file system. Subtract PgspOut from this value to get the number of 4K pages written to the file system per second during the monitoring interval.

Sios

The number of Iamp O requests made by the virtual memory manager per second during the monitoring interval.

Memory

Displays the distribution of real memory size and memory usage. The following data will be reported:

Real,MB

The size of real memory in MB.

% Comp

The percentage of real memory currently allocated to the calculation page frame. Calculating page frames are usually those supported by paging space.

% Noncomp

The percentage of real memory currently allocated to non-calculated page frames. Non-computational page frames are usually those supported by file space, which can be data files, executables, or shared library files.

% Client

The percentage of real memory currently allocated to cache files installed remotely.

Paging space

Displays the size and usage of paging space. The following data will be reported:

Size,MB

The sum of all paging space on the system, in MB.

% Used

The total percentage of paging space currently in use.

% Free

The total percentage of currently unused paging space.

NFS

Displays NFS statistics by calls per second. The following data will be reported:

Server V2 calls per second

Client V2 calls per second

Server V3 calls per second

Client V3 calls per second

Total WPAR

Displays the total number of workload partitions defined in the system. The total number of workload partitions that can be defined, active, interrupted, or transitional.

Active WPAR

Displays the total number of workload partitions for which the resource is active.

AME

Displays memory compression statistics on systems with Active Memory? extension enabled. The following data will be reported:

TMEM,MB

Real memory size in megabytes.

CMEM,MB

Compression pool size in megabytes.

EF [T/A]

Expansion factor: target and actual value.

CI

The number of page calls from the compression pool.

CO

The number of page calls out of the compression pool.

The variable section displayed by topas can have one, two, three, four, or five subsections. If more than one subsection is displayed, it is always displayed in the following order:

Processor usage

Network interface

Physical disk

File system

Workload Manager Class

Workload partition

Process

When the topas command starts, all subsections of the hot entity being monitored are displayed. The workload Manager (WLM) class subsection is displayed only when WLM is active.

WLM should be started to view WLM and WPAR statistics.

Tip: when there is no WPAR-specific information for the measure, the background will be called to display the system-wide value for the measure (that is, white text and black environment).

The following table provides the details of the subsections displayed by the topas command:

Project

Description

Processor usage

This section shows a bar chart to illustrate the usage of the cumulative processor. If there is more than one processor, you can display a list of processors by pressing c twice. Pressing c only once will close this section. The following fields are displayed in two formats:

User%

The percentage of processors used by programs running in user mode. (sort by default by percentage of users)

Kern%

The percentage of processors used by programs running in kernel mode.

Wait%

The percentage of time spent waiting for Iripple O.

Idle%

The percentage of idle time of the processor.

Physc

The number of physical processors consumed. Appears only when running a partition using a shared processor.

% Entc

Percentage of authorized capacity consumed. Appears only when running a partition using a shared processor.

When this subsection first displays a list of heat processors, the list is sorted by the User% field. However, you can sort the list by other fields by moving the cursor to the top of the desired column.

Network interface

This section shows a list of hot network interfaces. The maximum number of interfaces displayed is the number of hot interfaces being monitored, as specified by the-n flag. If other subsections are also being displayed, a smaller number of interfaces will be displayed. Press the n key to close this section. Press the n key again to display a report summary of all network interface activities in a row. Both reports display the following fields:

Interf

The name of the network interface.

BPS

The total throughput per second during the monitoring interval in kilobytes. The value of this field is the sum of the kilobytes received per second and the kilobytes sent.

I-Pack

The number of packets received per second during the monitoring interval.

O-Pack

The number of packets sent per second during the monitoring interval.

KB-In

The amount of data received per second during the monitoring interval in kilobytes.

KB-Out

The amount of data sent per second during the monitoring interval, in kilobytes.

When this subsection first displays the list of hot network interfaces, sort the list by the BPS field. However, you can sort the list by other fields by moving the cursor to the top of the desired column. Sorting is valid only for up to 16 network adapters.

Physical disk

This section displays a list of thermophysical disks. The maximum number of physical disks displayed is the number of thermophysical disks being monitored (as specified with the-d flag). If other subsections are also being displayed, a smaller number of physical disks will be displayed. Press d to close this section. Press d again to display a report summary of all physical disk activity in a row. Both reports display the following fields:

Disk

The name of the physical disk.

Busy%

The percentage of time that the physical disk is active (bandwidth usage of the drive).

BPS

The amount of data (in kilobytes) transferred (read and write) per second during the monitoring interval. The value of this field is the sum of the KB-Read value and the KB-Writ value.

TPS

The number of transfers sent to the physical disk per second. The transfer is an Ithumb O request to the physical disk. Multiple logical requests can be combined into a single Ipicuro request to the disk. The transmission size is uncertain.

KB-Read

The amount of data, in kilobytes, read from the physical disk per second.

KB-Writ

The amount of data written to the physical disk per second, in kilobytes.

When the list of thermophysical disks is first displayed in this section, sort the list by the BPS field. However, you can sort the list by other fields by moving the cursor to the top of the desired column. Sorting is valid only for up to 128 physical disks.

File system

This section displays a list of hot file systems. If the file system subsection is displayed, the physical disk subsection is not displayed. The maximum number of file systems displayed is the number of hot file systems mirrored (when they are specified by the-f flag). If other subsections are also displayed, fewer file systems will be displayed. To close this section, press d. When you press the f key, displays a summary of the active single-line report for all file systems. When you press the f key again, the file system subsection is no longer displayed and the physical disk subsection is displayed. Both reports display the following fields:

File system

The name of the file system.

BPS

The amount of data (in kilobytes) transferred (read and write) per second during the monitoring interval. The value of this field is the sum of the KB-Read field value and the KB-Writ field value.

TPS

The number of transfers sent to the file system per second. The transfer is an Iswap O request to the file system. Multiple logical requests can be combined into a single Ipicuro request into the file system. The size of the transmission is uncertain.

KB-Read

The amount of data read from the file system per second, in kilobytes.

KB-Writ

The amount of data written to the file system per second, in kilobytes.

When this subsection displays a list of file systems, sort the list by the BPS field. However, you can sort using other fields by moving the cursor to the top of the target column.

Tip: if the file system name exceeds the width of the displayed field, the file name will be truncated. Truncation includes the first and last characters of the file system, and the middle part of the name is replaced with a period (..). For example, if the file system name is filesystem001234, the name will be displayed as files..01234.

WLM class

This section 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 by the-w flag. If other subsections are also being displayed, a smaller number of WLM classes will be displayed. Press w to close this section. The following fields are displayed for each class:

% processor Utilization

The average processor utilization of the WLM class during the monitoring interval.

% Mem Utilization

The average memory usage of the WLM class during the monitoring interval.

% Blk I PUBO

The average percentage of block iUnix O of the WLM class during the monitoring interval.

When this subsection first displays a list of hot WLM classes, the list is sorted by the CPU% field. However, you can sort the list by other fields by moving the cursor to the top of the desired column.

Tip: if the WLM class name exceeds the displayed field width, the WLM class name will be truncated. Truncation includes the first and last characters of the WLM class, and the middle part of the name is replaced with a period (..). For example, if the WLM class name is unclassified00123, then the WLM class name will be displayed as uncla..00123.

Workload partitions

If called with the-@ flag, the workload partition subsection replaces the WLM subsection. This section displays a list of hot workload partitions. The maximum number of workload partitions displayed is the number of monitored hot WPAR (when they are specified using the-w-@ flag). If other subsections are also displayed, fewer WPAR will be displayed. To close the workload partition subsection, press the @ key. The following fields are displayed for each WPAR:

WPAR

The name of the workload partition (WPAR).

% processor Utilization

The average processor utilization of WPAR during the monitoring interval.

% Mem Utilization

The average memory usage of WPAR during the monitoring interval.

% Blk I PUBO

The average percentage of block I _ peg O of the WPAR during the monitoring interval.

When this subsection displays a list of hot WPAR, the list is sorted by the CPU% field. However, you can use other fields to sort by moving the cursor to the top of the target column that you want to sort the list.

Tip: if the WPAR name exceeds the displayed field width, the WPAR name will be truncated. Truncation includes the first and last characters of the WPAR class, and the middle part of the name is replaced with a period (..). For example, if the name of WPAR is neptune00123, then WPAR will be displayed as neptu..00123.

Process

This section displays a list of hot processes. The maximum number of processes displayed is the number of hot processes being monitored, as specified by the-p flag. If other subsections are also being displayed, a smaller number of processes will be displayed. Press p to close this section. Processes are sorted by the usage of their processors during the monitoring interval. The following fields are displayed for each process:

Name

The name of the executable program executed in the process. The name has been stripped of any pathname and parameter information and truncated to a length of 9 characters.

Process identification

The process identity of the process.

Percentage of CPU usage

The average processor utilization of the process during the monitoring interval. When a process is first displayed, this value is the average processor utilization over the lifetime of the process.

Paging space used

The size of the paging space allocated to this process. This can be seen as a representation of the process coverage area, but does not include the memory used to maintain the executable program and any shared libraries it depends on.

Process owner (if the WLM section is closed)

The user name of the user who owns this process.

Workload Manager (WLM) class (if the WLM section is open)

The WLM class to which the process belongs.

WPAR (if the WPAR section is open)

The WPAR name to which the process belongs.

Tip: if the WLM class / WPAR name exceeds the displayed field width, the WLM class / WPAR name will be truncated. Truncation includes the first and last characters of the WLM class / WPAR, and the middle part of the name will be replaced with a period (..). For example, if the WLM class / WPAR name is unclassified00123, then the WLM class / WPAR name will be displayed as uncla..00123.

Adapter panel view

When using the topas-D command, press d to display the Adapter panel view. In this panel, the following measures are displayed:

Project

Description

Adapter

The name of the adapter.

KBPS

The amount of data (in kilobytes) transferred (read or write) in the adapter per second.

TPS

Indicates the average number of transmissions made by the adapter per second.

KB-R

The total amount of data read from the adapter in kilobytes.

KB-W

The total amount of data written to the adapter in kilobytes.

If you press the f key, the following details of the disks belonging to the adapter are displayed on the adapters panel:

Project

Description

Vtargets/Disk

The name of the virtual target device or disk.

Busy%

The percentage of time that the virtual target device or disk is active (bandwidth usage of the virtual target device or disk).

KBPS

The amount of data, in kilobytes, read and written per second during the monitoring interval. The value of this field is the sum of the KB-R measure and the KB-W measure.

TPS

The number of transfers sent to the virtual target device or disk per second. A transfer is an Icano request to a virtual target device or disk. Multiple logical requests can be combined into a single Ipicuro request to the disk. The transmission is medium size.

KB-R

The amount of data read from the virtual target device or disk per second, in kilobytes.

KB-W

The amount of data written to the virtual target device or disk per second, in kilobytes.

AQD

The average number of requests waiting to be sent to the virtual target device or disk.

AQW

The average time that the transport request spent in the waiting queue. The suffix indicates the unit of time. The default time unit is milliseconds.

ART

The average time to receive a response from the supervisor server for a read request sent. The suffix indicates the unit of time. The default time unit is milliseconds.

AWT

The average time to receive a response from the supervisor server for a write request sent. The suffix indicates the unit of time. The default time unit is milliseconds.

MRT

For read requests sent, the maximum time to receive a response from the supervisor server. The suffix indicates the unit of time. The default time unit is milliseconds.

MWT

The maximum time to receive a response from the supervisor server for a write request sent. The suffix indicates the unit of time. The default time unit is milliseconds.

Virtual Adapter Panel View

When you run the topas-D command, press v to display the Virtual Adapter panel view. In this panel, the following measures are displayed:

Project

Description

VAdapter

The name of the virtual adapter.

TPS

The number of transmissions sent to the adapter per second.

KBPS

The amount of data (in kilobytes) transferred (read or write) in the adapter per second.

KB-R

The number of blocks received by the adapter from the supervisor server per second.

KB-W

The number of blocks sent to the supervisor server from this adapter per second.

AQD

The average number of requests waiting to be sent to the adapter.

AQW

The average time that the transport request spent in the waiting queue. The suffix indicates the unit of time. The default time unit is milliseconds.

ART

The average time to receive a response from the supervisor server for a read request sent. The suffix indicates the unit of time. The default time unit is milliseconds.

AWT

The average time to receive a response from the supervisor server for a write request sent. The suffix indicates the unit of time. The default time unit is milliseconds.

MRT

For read requests sent, the maximum time to receive a response from the supervisor server. The suffix indicates the unit of time. The default time unit is milliseconds.

MWT

The maximum time to receive a response from the supervisor server for a write request sent. The suffix indicates the unit of time. The default time unit is milliseconds.

If you press the f key, the following details of the disks that belong to the adapter are displayed on the virtual adapters panel

Project

Description

Vtargets/Disk

The name of the virtual target device or disk.

Busy%

The percentage of time that the virtual target device or disk is active (bandwidth usage of the virtual target device or disk).

KBPS

The amount of data, in kilobytes, read and written per second during the monitoring interval. The value of this field is the sum of the KB-R measure and the KB-W measure.

TPS

The number of transfers sent to the virtual target device or disk per second. A transfer is an Icano request to a virtual target device or disk. Multiple logical requests can be combined into a single Ipicuro request to the disk. The transmission is medium size.

KB-R

The amount of data read from the virtual target device or disk per second, in kilobytes.

KB-W

The amount of data written to the virtual target device or disk per second, in kilobytes.

AQD

The average number of requests waiting to be sent to the virtual target device or disk.

AQW

The average number of requests waiting (reported by request, in milliseconds). The suffix indicates the unit of time. The default time unit is milliseconds.

ART

The average time to receive a response from the supervisor server for a read request sent. The suffix indicates the unit of time. The default time unit is milliseconds.

AWT

The average time to receive a response from the supervisor server for a write request sent. The suffix indicates the unit of time. The default time unit is milliseconds.

MRT

For read requests sent, the maximum time to receive a response from the supervisor server. The suffix indicates the unit of time. The default time unit is milliseconds.

MWT

The maximum time to receive a response from the supervisor server for a write request sent. The suffix indicates the unit of time. The default time unit is milliseconds.

MPIO panel view

When using the topas-D command, press m to display the MPIO panel view. In this panel, the previous section contains the same metrics as the disks panel.

The next section of the panel contains the following fields:

Project

Description

Path

The name of the path.

Busy%

The percentage of time that the path is active (the bandwidth usage of the path).

KBPS

The amount of data, in kilobytes, read and written per second during the monitoring interval. The value of this field is the sum of the KB-R measure and the KB-W measure.

TPS

The number of transmissions sent out in this path per second.

KB-R

The amount of data, in kilobytes, read in this path per second.

KB-W

The amount of data, in kilobytes, written in this path per second.

Panel freeze

The spacebar on the keyboard acts as a switch to freeze the topas panel. If frozen, topas stops data collection and continues to display data from previous iterations. You can move on the panel and sort the data according to the selected column. In a frozen state, if you move between panels, some panels may not display this data. In this case, press the Spacebar to thaw the topas panel.

Rolling

If the amount of data exceeds the size of the topas window, the page up and page down keys are used to scroll between the data. The data is sorted according to the selected column.

Note: the above functions are provided with the selected panel (via topas).

Icano memory usage Pool Panel

If the logical partitioning panel (topas-L) is enabled in shared memory mode, you can press the e key to display the "IWeiO memory usage pool" panel.

The following metrics are displayed in the next section of this panel:

Project

Description

Iompn

The name of the Icano memory pool.

Iomin

The minimum Imax O memory usage of the pool.

Iodes

The expected Imax O memory usage of the pool.

Ioinu

The current Imax O memory usage of the pool.

Iores

Reserved Imax O memory usage for the pool.

Iohwm

The maximum Istroke O memory usage (high water mark) used by the pool.

Ioafl

For this pool, allocate the total number of failed requests.

Cross-partition views and records

For all AIX partitions that can be identified as belonging to the same hardware platform, it displays metrics similar to the lparstat command. Dedicated and shared partitions and their corresponding measures are displayed in separate sections. The top section represents the aggregated data collected from the partition set to show the entire partition, memory, and processor activity.

To enable this panel remotely for collection from other partitions, you need to use the latest updates from perfagent.tools and bos.perf.tools to support this feature. For earlier versions of AIX, the topas command also collected remote data from the partition where the Performance Aide product (perfagent.server) was installed. The topas-C command may not find partitions on other subnets. To avoid this, create a $HOME/Rsi.hosts file that contains the standard host name for each partition, including the domain, and a host occupies one line.

Note: the topas-C command sends broadcast packets to all logical partitions (LPAR) within the same subnet, but only handles responses from LPAR in the same CEC.

The following measures are displayed in the initial cross-partition panel. Other measures with a full description label can be displayed by using the key toggles identified in the other Cross-Partition Panel subcommands section.

Total number of partitions:

Project

Description

Shr

Number of shared partitions based on the system processor.

Ded

The number of dedicated partitions based on the system processor.

Memory (in GB)

Project

Description

Mon

Total memory of the monitored partition

Avl

Available memory of the partition set

InUse

Memory in use on the monitored partition

Processor:

Project

Description

Shr

Number of shared processors

Ded

Number of dedicated processors

PSz

Number of active physical processors in the pool of physical shared processors used by this LPAR

APP

The number of physical processors available in the shared pool. This is equal to the idle cycle pool reported as the number of processors

Don

Total number of processors contributed to the pool

Shr_PhysB

Total number of physical processors busy for all shared partitions

Ded_PhysB

Total number of physical processors busy for all dedicated partitions

Single partition data:

Project

Description

Host

Hostnam

OS

Operating system level

Mod

The way each partition. This method is displayed as 3 characters.

Character

The first character indicates the CPU in the partition. The second character indicates how the partition is in memory. The third character indicates the energy consumption status of the partition.

Mem

Total memory (in gigabytes).

InU

The amount of memory in use (in gigabytes).

Lp

Number of logical processors.

Us

The percentage of processors used by programs executed in user mode.

Sy

The percentage of processors used by programs executed in kernel mode.

Wa

The percentage of time spent waiting for Iripple O.

Id

The percentage of time that the processor was idle.

PhysB

The number of physical processors that are busy.

Ent

Authorized usage (shared mode only).

% Entc

The percentage of usage consumed (shared only).

Vcsw

Average number of virtual context switches per second (shared mode only).

PhI

The average number of Phantom read interruptions per second (shared mode only).

Pmem

Physical memory that provides support for partitioned logical memory (if in shared memory mode).

% idon

The percentage of physical processors used when explicitly contributing idle cycles. This measure applies only to contribution-specific partitioning.

% bdon

The percentage of physical processors used when contributing busy cycles. This measure applies only to contribution-specific partitioning.

% istl

The percentage of physical processors used when the hypervisor is stealing idle cycles. This measure applies only to dedicated partitioning.

% bstl

The percentage of physical processors used when the hypervisor is stealing busy cycles. This measure applies only to dedicated partitioning.

For shared partitions:

The first character

Description

C

Enable and restrict SMT

C

Disable and restrict SMT

U

Enable and do not restrict SMT

U

Disable and do not restrict SMT

For dedicated partitions:

The first character

Description

S

Enable and do not contribute SMT

D

Disable and contribute SMT

D

Enable and contribute SMT

-

Disable and do not contribute SMT

Second character

Description

M

Enable AMS, but disable AME

-

Disable both AME and AMS

E

Enable both AME and AMS

E

Enable AME, but disable AMS

The third character

Description

S

Static power saving mode is enabled

D

Power saving mode has been disabled

D

Dynamic power saving mode is enabled

-

Unknown / undefined

E

Power saving mode enabled

D

Power saving mode has been disabled

When no dedicated partition is contributed, the% idon and% bdon metrics are not displayed.

Requirement: at least one partition to monitor must configure Pool usage permissions (PUA) for the pool information metrics to be collected.

For cross-partition monitoring / logging, some global data for some partitions is not available. The-o option allows you to specify these fields on the command line. (optional) you can configure the system to allow the topas command to query HMC directly for this information. This requires the following steps:

1. Install OpenSSH on the partition.

two。 Enable remote command support for user hscroot on HMC to allow ssh connections to be opened from the partition.

3. Configure ssh on HMC so that the password of the HMC user hscroot is not required when querying from the selected partition. This requires .ssh / authorized_keys2 on HMC for the user to log in to hscroot.

4. Run ssh-l hscroot hmc_address date from the partition to confirm that the date can be displayed without entering a password.

5. When you run the topas command, use the topas-o option described in the usage table to specify the managed system and HMC name.

Limitation: this feature is currently available only for HMC v5 and above, and can only be enabled after careful consideration of security implications.

The following is displayed when the g key is pressed on the splash screen (a cross-partition view with a detailed title):

Topas CEC monitor interval: October Monday 22 00:08:00 2007

Partition Information memory (GB) processor virtual pool: 2

Monitored: 2 monitored: 6.2 monitored: 2.0 available pool processor: 5

Unmonitored:-shared physical busy: 0.00

Share: 0 available:-available:-dedicated physical busy: 0.05

Infinite: 0 unallocated:-unallocated:-contribute physical processor: 0.00

Restricted: 2 used: 1.9 share: 0 embezzlement of physical processor: 0.01

Dedicated: 2 dedicated: 2 (system) management program

Contribution: 0 contribution: 0 virtual context switching times: 347

Pool size: 0 Phantom read 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 20 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 00 0 99 0.00 177 0.1 0.0 0.0 0.0

Ptoolsl3 A54 S 3.1 0.9 2 00 0 99 0.00 170 0.2 0.0 0.0 0.0

The following headings are on the previous screen:

Partition information:

Project

Description

Monitored

Number of partitions monitored

Unmonitored

Number of partitions that are not monitored

Shared

Number of shared partitions

Uncapped

Number of unrestricted shared partitions

Capped

Restricted number of shared partitions

Dedicated

Number of dedicated partitions

Donating

Number of partitions currently being contributed

Memory:

Project

Description

Monitored

Total monitored memory

UnMonitored

Total memory unmonitored

Available

Total available memory

UnAllocated

Total memory not allocated to any partition

Consumed

Total memory consumed by the partition

Processor:

Project

Description

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 assigned to any partition

Shared

Number of processors in the shared partition

Dedicated

Number of processors in dedicated partition

Donated

The sum of the processors of all contributing partitions

Pool Size

Number of active shared processors in all pools

Avail Proc Pool

Number of physical processors in the available pool

Shr Physical Busy

The sum of busy physical partitions in all shared partitions

Ded Physical Busy

The sum of busy private partitions in all dedicated partitions

Donated Phys. Processors

The sum of the contributed processor cycles reported to the number of processors in all partitions

Stolen Phys. Processors

The sum of embezzled processor cycles reported as processors in all partitions

Virtual Pools

Number of virtual pools

Virt. Context Switch

The number of virtual context switches per second during the monitoring interval.

Phantom Interrupts

Total Phantom read interruptions per second during the monitoring interval

When the topas command runs within any cross-partition view, press p to launch the pool panel. The following is an example, which is shown as:

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

You can scroll up and down in the pool identity column and press f to list only the shared partitions that belong to the poolid where the cursor is located. The following headings may appear on the screen:

Project

Description

Psize

Effective maximum capacity of the pool

Entc

Authorized capacity of the pool

Maxc

Maximum capacity of the pool

Physb

The sum of physical processors that are busy in the shared partition of the pool

App

Available physical processors in the pool

Mem

The sum of monitored memory for all shared partitions in the pool

Muse

The sum of the consumed memory of all shared partitions in the pool

When the topas command is running in any cross-partition view, press the v key to display the virtual Imax O server / client throughput panel. The following measures are displayed:

Project

Description

Server

The name of the VIO server.

Client

The name of the VIO client.

KBPS

The amount of data, in kilobytes, read and written per second during the monitoring interval. The value of this field is the sum of the KB-R measure and the KB-W measure.

TPS

The number of transmissions sent per second.

KB-R

The amount of data read per second in kilobytes.

KB-W

The amount of data written per second, in kilobytes.

AQD

Average number of requests waiting to be sent.

AQW

The average number of requests waiting (reported by request, in milliseconds). The suffix indicates the unit of time. The default time unit is milliseconds.

ART

The average time to receive a response from the supervisor server for a read request sent. The suffix indicates the unit of time. The default time unit is milliseconds.

AWT

The average time to receive a response from the supervisor server for a write request sent. The suffix indicates the unit of time. The default time unit is milliseconds.

MRT

For read requests sent, the maximum time to receive a response from the supervisor server. The suffix indicates the unit of time. The default time unit is milliseconds.

MWT

The maximum time to receive a response from the supervisor server for a write request sent. The suffix indicates the unit of time. The default time unit is milliseconds.

If the topas command is running in the Virtual Igamot O Server / client Throughput panel, after selecting the server from the Virtual Igamot O Server / client Throughput panel, press d to switch to the VIO server / client disk details panel. The previous section of this panel displays the server adapter details, while the next section shows the target device and client disk details. To list the target devices and client disks that belong to the adapter, select the adapter and press f.

The Virtual Icano Server / client disk details panel displays the following metrics:

Project

Description

Adapter

The name of the server adapter.

Vtargets

The name of the virtual target device that belongs to the server adapter.

Client_disk

The name of the client disk that has been mapped to the virtual target device of the server adapter.

The previous section of the panel displays the following details of the adapter:

Project

Description

KBPS

The amount of data (in kilobytes) transferred (read or write) in the adapter per second.

TPS

The number of transmissions sent to the adapter per second.

KB-R

The total amount of data read from the adapter in kilobytes.

KB-W

The total amount of data written to the adapter in kilobytes.

AQD

The number of requests waiting to be sent to the adapter.

AQW

The time that the transport request spent waiting in the queue. The suffix indicates the unit of time. The default time unit is milliseconds.

ART

For the read request sent, the time to receive the response from the supervisor server. The suffix indicates the unit of time. The default time unit is milliseconds.

AWT

For the write request sent, the time to receive the response from the supervisor server. The suffix indicates the unit of time. The default time unit is milliseconds.

MRT

For read requests sent, the maximum time to receive a response from the supervisor server. The suffix indicates the unit of time. The default time unit is milliseconds.

MWT

The maximum time to receive a response from the supervisor server for a write request sent. The suffix indicates the unit of time. The default time unit is milliseconds.

The panel displays the following details for the virtual target device and client disk:

Project

Description

Busy%

The percentage of time that the virtual target device or disk is active (bandwidth usage of the virtual target device or disk).

KBPS

The amount of data, in kilobytes, read and written per second during the monitoring interval. The value of this field is the sum of the KB-R measure and the KB-W measure.

TPS

The number of transfers sent to the virtual target device or disk per second. A transfer is an Icano request to a virtual target device or disk. Multiple logical requests can be combined into a single Icano request to a virtual target device or disk. The transmission is medium size.

KB-R

The amount of data read from the virtual target device or disk per second, in kilobytes.

KB-W

The amount of data written to the virtual target device or disk per second, in kilobytes.

AQD

The average number of requests waiting to be sent to the virtual target device or disk.

AQW

The average number of requests waiting (reported by request, in milliseconds). The suffix indicates the unit of time. The default time unit is milliseconds.

ART

The average time to receive a response from the supervisor server for a read request sent. The suffix indicates the unit of time. The default time unit is milliseconds.

AWT

The average time to receive a response from the supervisor server for a write request sent. The suffix indicates the unit of time. The default time unit is milliseconds.

MRT

For read requests sent, the maximum time to receive a response from the supervisor server. The suffix indicates the unit of time. The default time unit is milliseconds.

MWT

The maximum time to receive a response from the supervisor server for a write request sent. The suffix indicates the unit of time. The default time unit is milliseconds.

To display the memory pools panel from the CEC panel, press m. This panel displays statistics for all memory pools in the system. To display the partition corresponding to the pool in the next section of the panel, select a specific memory pool and press f.

The title section of the panel displays the following values:

Project

Description

Mshr

The number of logical partitions (LPAR) running in shared memory mode.

Mded

The number of LPAR running in dedicated memory mode.

Pools

The total number of memory pools in the system.

Mpsz

The total amount of physical memory in gigabytes for all memory pools.

MPuse

The total memory in gigabytes used by the LPAR associated with all pools.

Entl

The total LPAR O memory usage (in gigabytes) for all the pools.

Use

The total LPAR O memory usage (in gigabytes) used by all Icancano in all pools.

Mon

The total monitored memory of the system (the sum of the Mpsz metric and the total memory metric of the dedicated memory partition).

InUse

The total memory being used by the system (the sum of the MPuse metric and the total memory metric being used by the dedicated memory partition).

Avl

The total available memory of the system (the difference between Mon metrics minus InUse metrics).

The following values for the pool are displayed:

Project

Description

Mpid

Identification of the memory pool.

Mpsz

The total physical memory size of the memory pool in gigabytes.

Mpus

The total memory of the memory pool being used (the sum of the physical memory allocated to all LPAR in the pool).

Mem

The size in gigabytes of aggregate logical memory for all partitions in the pool.

Memu

Aggregate logical memory (in gigabytes) used by all partitions in the pool.

Iome

The aggregate I LPAR O memory usage, in gigabytes, configured for all the pool's users.

Iomu

The amount of aggregated Iswap O memory usage (in gigabytes) used by all LPAR in the pool.

Hpi

The aggregate number of hypervisor page fault failures that have occurred for all LPAR in the pool.

Hpit

The aggregate time (in milliseconds) spent by all LPAR in the pool waiting for the hypervisor page to call in.

The following values for partitions in the pool are displayed:

Project

Description

Mem

The logical memory size of the partition in gigabytes.

Memu

The logical memory (in gigabytes) used by the partition.

Meml

Logical memory lent to the hypervisor by LPAR.

Pmem

The physical memory in gigabytes allocated to the partition from the memory pool.

Iom

The amount of memory (in gigabytes) that is configured for LPAR.

Iomu

The amount of memory used by LPAR in gigabytes.

Hpi

The number of missing page faults in the hypervisor.

Hpit

The amount of time, in milliseconds, that the hypervisor waited for the page to be called.

Vcsw

The average number of virtual context switches per second.

Physb

A physical processor that is busy.

% entc

Percentage of processor usage consumed.

Cluster utilization view

A cluster is a group of related partitions or nodes. The Cluster Utilization view can show the utilization of a HA cluster or a user-defined cluster. For all AIX partitions that can be identified as belonging to the same hardware platform, it displays metrics similar to the lparstat command. Dedicated and shared partitions and their corresponding measures are displayed in separate sections. The top section represents the aggregated data collected from the partition set to show the entire partition, memory, and processor activity.

The initial cluster utilization panel displays the following metrics. Other metrics with a full description label can be displayed by using the key toggles identified in other cluster utilization dashboard command topics.

Total number of partitions:

Project

Description

Shr

Number of shared partitions based on the system processor.

Ded

The number of dedicated partitions based on the system processor.

Memory (in GB):

Project

Description

Mon

The total memory of the monitored partition.

InUse

The memory being used on the monitored partition.

Processor:

Project

Description

Shr

Number of shared processors.

Ded

Number of dedicated processors.

Shr_PhysB

The total number of physical processors that are busy for all shared partitions.

Ded_PhysB

The total number of physical processors that are busy for all dedicated partitions.

Single partition data:

Project

Description

Host

Hostname.

CEC

CEC logo.

OS

Operating system level

Mem

Total memory (in gigabytes).

M

The way each partition.

InU

The amount of memory in use (in gigabytes).

Lp

Number of logical processors.

Us

The percentage of processors used by programs executed in user mode.

Sy

The percentage of processors used by programs executed in kernel mode.

Wa

The percentage of time spent waiting for Iripple O.

Id

The percentage of time that the processor was idle.

PhysB

The number of physical processors that are busy.

Ent

Authorized usage (shared mode only).

% Entc

The percentage of usage consumed (shared only).

Vcsw

Average number of virtual context switches per second (shared mode only).

For shared partitions

Character

Description

C

Enable and restrict SMT

C

Disable and restrict SMT

U

Enable and do not restrict SMT

U

Disable and do not restrict SMT

For dedicated partitions

Character

Description

S

Enable and do not contribute SMT

D

Disable and contribute SMT

D

Enable and contribute SMT

-

Disable and do not contribute SMT

The following data is displayed when you press g on the splash screen, which generates the Cluster Utilization view with a detailed title:

Topas CEC Cluster Monitor ID: Interval: 10 Thu Apr 2 16:13:18 2009

Partitions Memory (GB) Processor

Shr: 2 Mon: 6.0 Shr: 1.5 Shr_PhyB: 0.01

Ded: 2 InU: 3.0 Ded: 2 Ded_PhyB: 0.00

Host CEC OS M Mem InU Lp Us Sy Wa Id PhysB Vcsw Ent EntC

-- shared--

Clock16 19318230 A61 U 2.0 1.1 2 00 0 99 0.00 423 0.75 0.6

Clock15 19318230 A61 U 2.0 1.6 2 0 0 0 99 0.01 985 0.75 0.9

Host CEC OS M Mem InU Lp Us Sy Wa Id PhysB Vcsw

-- dedicated-

Ses10 19318230 A61 D 2.0 1.1 20 00 99 0.00 0

Clock10 19318230 A61 D 0.0 0.0 20 00 99 0.00 742

The following display when press g key from the above panel

Which brings the cluster utilization view with detailed headers:

Topas Cluster Monitor ID: Interval: 10 Thu Apr 2 16:13:44 2009

Partition Info Memory (GB) Processor Supplier: ses10.in.ibm.com

Monitored: 4 Monitored:6.0 Monitored: 3.5 Shr Physical Busy: 0.01

Shared: 2 Consumed: 3.0 Shared: 1.5 Ded Physical Busy: 0.00

Uncapped: 2 Dedicated: 2

Capped: 2

Dedicated: 2

Host CEC OS M Mem InU Lp Us Sy Wa Id PhysB Vcsw Ent EntC

-- shared--

Clock16 19318230 A61 U 2.0 1.1 2 00 0 99 0.00 423 0.75 0.6

Clock15 19318230 A61 U 2.0 1.6 2 0 0 0 99 0.01 985 0.75 0.9

Host CEC OS M Mem InU Lp Us Sy Wa Id PhysB Vcsw

-- dedicated-

Ses10 19318230 A61 D 2.0 1.1 20 00 99 0.00 0

Clock10 19318230 A61 D 0.0 0.0 20 00 99 0.00 742

Implementation details

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 making changes to the disk and network adapters.

Mark

Project

Description

-@ wparname

Displays measures specific to-WPAR. If you specify WPAR with the wparname parameter, then topas will monitor the WPAR.

-chotprocessor

Use the hotprocessor parameter to specify the number of heat processors to monitor. This is also the maximum number of processors displayed when there is enough free space on the screen. If this number exceeds the number of available processors, only installed processors will be monitored and displayed. If this parameter is omitted, 2 is assumed to be the default. If you specify a value of 0 (zero), 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. These measures are similar to those collected by the lparstat command. Dedicated and shared partitions are displayed, and a set of aggregation values outlines the entire hardware system partition set. You can use the line command to set certain values that can only be used on the HMC platform (if the HMC connection is not available):

-G

Displays the Cluster Utilization panel. The topas command collects a set of metrics from AIX partitions running on the same hardware platform. These measures are similar to those collected by the lparstat command. Dedicated and shared partitions are displayed.

-D

Displays the disk Metrics screen (disk panel view). This screen reports disk service time, disk queuing metrics, and disk throughput. Report the following metrics:

Disk

The name of the physical disk.

Busy%

The percentage of time that the physical disk is active (bandwidth utilization of the disk).

KBPS

The amount of data, in kilobytes, read and written per second during the monitoring interval. The value of this field is the sum of the KB-R measure and the KB-W measure.

TPS

The number of transfers sent to the physical disk per second. The transfer is an Ithumb O request to the physical disk. Multiple logical requests can be combined into a single Ipicuro request to the disk. The transmission is medium size.

KB-R

The amount of data, in kilobytes, read from the physical disk per second.

ART

The average time to receive a response from the supervisor server for a read request sent. The suffix indicates the unit of time. The default time unit is milliseconds.

MRT

For read requests sent, the maximum time to receive a response from the supervisor server. The suffix indicates the unit of time. The default time unit is milliseconds.

KB-W

The amount of data written to the physical disk per second, in kilobytes.

AWT

The average time to receive a response from the supervisor server for a write request sent. The suffix indicates the unit of time. The default time unit is milliseconds.

MWT

The maximum time to receive a response from the supervisor server for a write request sent. The suffix indicates the unit of time. The default time unit is milliseconds.

AQW

The average number of requests waiting (reported by request, in milliseconds).

AQD

The average number of requests waiting to be sent to disk.

With the-D flag specified, you can run the following subcommands:

To view the adapter panel, press d.

To display all virtual adapters (virtual adapter panels) that exist in the partition, press the v key.

To display the disks that belong to the adapter or virtual adapter, press f.

To display the MPIO panel, press m. This panel displays disk details and path details. To list the path to the disk, press f.

Restrictions:

The-D option provides a disk panel view to report disk service time, disk queuing metrics, and disk throughput. Whenever the-D option is enabled, the disk minimum and maximum service time metrics are reset during the first interval. Because the service time metric is reset during the first interval of the-D option, it affects the use of the disk service time metric by an existing instance of the-D option or by some other consumer.

-d hotdisk

Specifies the number of disks to be monitored. The hotdisk parameter specifies the number of hot disks to monitor. This is also the maximum number of disks to display when there is enough free space on the screen. When this number exceeds the number of disks installed, only the installed disks will be monitored and displayed. If this parameter is omitted, 2 is assumed to be the default. If the specified value is 0 (zero), disk information is not monitored.

-E

Displays statistics for shared Ethernet adapters on the virtual Imap O server. The following measures are displayed:

KBPS

The total throughput per second during the monitoring interval in kilobytes. The value of this field is the sum of the kilobytes received per second and the kilobytes sent.

I-Pack

The number of packets received per second during the monitoring interval.

O-Pack

The number of packets sent per second during the monitoring interval.

KB-In

The amount of data received per second during the monitoring interval in kilobytes.

KB-Out

The amount of data sent per second during the monitoring interval, in kilobytes.

-F

Displays the file system display. When you specify the-@ flag or the @ subcommand, the file system is displayed in two windows. The top of the monitor displays a list of active WPAR. This table can be sorted by any column. Displays the reported file system service time, file system queuing metric, and file system throughput. Report the following metrics:

File system

The name of the file system.

KBPS

The amount of data transferred (read and write) per second during the monitoring interval. The value of this field is the sum of the KB-Read value and the KB-Writ value.

TPS

The number of transfers sent to the file system per second. The transfer is an Iswap O request to the file system. Multiple logical requests can be combined into a single Ipicuro request into the file system. The size of the transmission is uncertain.

KB-Read

The amount of data read from the file system per second, in kilobytes.

KB-Writ

The amount of data written to the file system per second, in kilobytes.

Open

The logical number of files opened.

Create

The logical number of file creation.

Lock

The number of files that are locked on the file system.

Tip: if the file system name exceeds the width of the displayed field, the file system name display will be truncated. Truncation includes the first and last characters of the file system, and the middle part of the name is replaced with a period (..). For example, if the file system name is filesystem001234, the file system name will be displayed as files..01234.

-f HotFS

Use the HotFS parameter to specify the number of file systems to monitor. This is also the maximum number of file systems displayed when there is enough space available. "if this number exceeds the number of available file systems, only mounted file systems are monitored and displayed." If the-f flag is not specified, the default value is two. If you specify a zero value, the file system information is monitored.

-h

Display 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-processors]

[- U username_owned_processes]

[- D |-P |-W |-L]

[- m]

-I interval

Sets the monitoring interval or recording interval (seconds). If you specify the-I flag with the interval parameter, the interval parameter sets the monitoring interval. The default value of the interval parameter is two seconds.

If you specify the-I flag with-R, the interval parameter becomes the recording interval for partition metrics. The default value of the interval parameter is 300 seconds. Valid values are 10, 15, 30, 60, 120, and 300 seconds.

-I remotepollinterval

For cross-partition screens, use the remotepollinterval parameter to set the sampling interval to collect data from remote partitions. The default value of the remotepollinterval parameter is 10 seconds. The allowed values are 10, 15, 30, 60, and 120 seconds.

-L

Displays the logical partitioning screen. This display reports similar data to the content provided to mpstat and lparstat.

In shared memory mode, this panel displays information about the Imax O memory usage of the partition. The existing% lbusy,% hypv, and hcalls measures are replaced with the following measures:

IOME

The Icano memory usage of the partition in gigabytes.

Iomu

The amount of Istroke O memory usage in gigabytes being used by the partition.

Pmem

Provides supported physical memory (in gigabytes) for the logical memory of the partition.

Hpi

The number of times the hypervisor page is called.

Hpit

The time in milliseconds to wait for the hypervisor page to be called.

When the-L flag is specified, you can press the e key to display the Imax O memory usage Pool panel. For more information about this panel, see the Imax O memory usage pool panel.

-M

Displays the memory Topology panel.

This screen reports data similar to the content provided by the lssrad command.

This panel consists of two sections:

The first part provides the memory topology from a SRAD perspective. At each REF1 system detail level, it provides each SRAD identity and the resources (memory and processor) associated with each identity.

The second part (that is, the CPU RAD screen) provides relevant data at the processor level.

The following measures are displayed in this panel:

REF1

The reference point provided by the first hardware is used to identify groups of resources that are adjacent to each other.

SRAD

Scheduler resource allocation domain identification.

TOTALMEM

The total amount of memory in SRAD (in MB).

INUSE

The amount of memory in use in the SRAD.

FREE

The amount of memory available in the SRAD.

FILECACHE

The number of file cache bytes consumed by the LRU daemon.

HOMETHRDS

The number of threads with SRAD as the home. Threads typically run on the CPU included in home SRAD, but this is not guaranteed. When you create a thread, the system selects home SRAD for that thread. You can change the home SRAD of a thread during its lifetime.

CPUS

The processor associated with this SRAD. 0 indicates that the cpu0 is associated with the corresponding SRAD identity. 0-28 means that all cpus from cpu0 to cpu28 is associated with the corresponding SRAD. If the CPU identities are not adjacent, the values are separated by commas.

TOTALDISP

The total number of threads dispatched from the corresponding processor during the interval.

LOCALDISP%

The percentage of threads dispatched locally within this SRAD (usually at the chip level).

NEARDISP%

The percentage of threads dispatched to CPU that are not local but are not far apart. Typically, these may be resources that share the same hardware node.

FARDISP%

The percentage of threads assigned to a processor that is usually located outside the hardware node.

Note: "local", "nearby" and "remote" have different hardware meanings for different architectures.

-m

Display in monochrome (no color).

-n hotni

Use the hotni parameter to specify the number of hot network interfaces to monitor. This is also the maximum number of network interfaces to display when there is enough free space on the screen. When this number exceeds the number of installed network interfaces, only the installed network interfaces will be monitored and displayed. If the argument is omitted, the default value of 2 will be used. If you specify a value of 0 (zero), no network information is monitored.

-P

Similar to the ps command, the-P flag displays the full-screen process screen. Displays a list of the busiest processes, similar to the process subsections when displayed by default, except that more columns are used to display more specification information for each process. This table can be sorted by any column. The following are the metrics displayed.

USER

The login name of the process owner.

PID

The process identity of the process.

PPID

The process identity of the parent process.

PRI

The priority of a process or kernel thread; a higher number indicates a lower priority.

NI

The process priority specified with the nice command is used to calculate the priority for other scheduled policies.

DATA RES

The real memory (resident set) data size of the process (4 KB pages).

TEXT RES

The real memory (resident set) text size of the process (4 KB pages).

PAGE SPACE

The size of the virtual working set used by the process (4 KB pages). Note: you cannot get the actual paging space allocation for each process using the topas command. For a more detailed report, see the svmon command.

TIME

The total execution time of the process.

CPU%

Percentage of processor utilization.

PGFAULTS

The number of Icano and other page faults.

COMMAND

Command name. Truncate the command name to 9 characters.

When you specify-@ (topas-P-@), the new field WPAR is displayed and the PPID field is removed. All other measures remain the same.

WPAR

The WPAR name to which the process belongs.

Tip: if the WPAR class name exceeds 12 characters and needs to be displayed in a 12-character format, two periods (..) will be added after the first five characters, followed by the last five characters. For example, if the WPAR class name is neptune001234, then the WPAR name is displayed as neptu..01234.

-photprocess

Use the hotprocess parameter to specify the number of heat processors to monitor. This is also the maximum number of processes to display when there is enough free space on the screen. If the argument is omitted, the default value of 20 will be used. If you specify a value of 0 (zero), no process information is monitored. The retrieval of process information constitutes the main overhead of topas. If process information is not required, always use this option to specify that process information is not required.

-t

Turn the tape screen portion on or off on the main topas screen.

-T

Displays the full-screen tape screen panel.

This panel displays the following measures:

Tape

The name of the tape device.

Busy%

Bandwidth utilization of the tape.

KBPS

The amount of data transferred (read or write) to tape per second (in kilobytes).

TPS

The average number of transfers sent to the tape per second.

KB-R

The total amount of data read from the tape in kilobytes.

ART

The average time to receive a response for a read request sent. The suffix indicates the unit of time. The default time unit is milliseconds.

MRT

For a read request sent, the maximum time to receive a response. The suffix indicates the unit of time. The default time unit is milliseconds.

KB-W

The total amount of data written to the adapter in kilobytes.

AWT

For a sent write request, the average time to receive the response. The suffix indicates the unit of time. The default time unit is milliseconds.

MWT

For a write request sent, the maximum time to receive a response. The suffix indicates the unit of time. The default time unit is milliseconds.

-U username

Used in conjunction with the-P flag, this flag shows the processes owned by the user specified using the username parameter. Only processes owned by the specified user can be displayed in All Process Display.

-V

Displays the Volume groups panel. The previous section of the panel reports the following measures for the volume group, while the next section reports the same measures for logical volumes.

LogicalVolume/VolumeGroup

Name of the logical volume or volume group.

TPS

The total number of Iwhite O requests during the interval in which the metric is displayed.

KB-R

The total amount of data read in kilobytes during the interval.

KB-W

The total amount of data written in kilobytes during the interval.

KBPS

The amount of data (in kilobytes) transferred (read or write) in a query logical volume or volume group per second.

-W

Displays a full-screen WLM screen, which is a split display. The top of the display displays a list of hot WLM classes, similar to the WLM class subsections in the default display, except that there is enough free space to display all the class names. This table can be sorted by any column.

If you specify the-@ flag or if you press the @ subcommand, the WPAR section is displayed instead of the WLM section. The WPAR section displays a list of hot WPAR. This table can be sorted by any column.

The bottom of the screen displays a list of the busiest processors (similar to the full-screen process screen), but only processes that belong to a WLM class or WPAR selected with the f key are displayed.

Note: if the WLM class is not active, the default system process will be displayed at the bottom of the screen.

-w [number of monitored hot WLM classes]

Use the hotwlmclass parameter to specify the number of hot workload manager (WLM) classes to monitor. This is also the maximum number of WLM classes to display when there is enough free space on the screen. When this number exceeds the number of installed WLM classes, only installed WLM classes will be monitored and displayed. If the argument is omitted, the default value of 2 will be used. If you specify a value of 0 (zero), no WLM class information is monitored.

Regular subcommand

When topas runs, it accepts single-character subcommands. At the end of each monitoring interval, the program checks one of the following subcommands and responds to the requested operation.

Project

Description

A

Displays all variable quantum nodes (processors, networks, disks, WLM, and processes) that are being monitored. Pressing the a key always returns the topas command to the initial main display screen.

C

Replace the current screen of the cumulative report with the processor subsection. When you press c again, the cumulative report is displayed. The number of busiest processors displayed depends on the available space on the screen.

C

Activate the Cross Partition panel. If the panel is currently active, the C key resets the panel to display the global summary, private, and shared sections. See the other cross-partition panel subcommands section below for options specific to this panel.

D

Replace the current screen of total disk activity with the list of busiest disks. When you press d again, the total disk activity is displayed. The number of busiest disks displayed depends on the available space on the screen.

D

Replace the current screen with the disk Metrics screen. This screen provides additional information about disk access time and disk queuing. If you press D again, the screen will switch back to the default home screen.

E

Displays the shared Ethernet Adapter panel in the VIO server.

F

Press f while moving the cursor over the WLM class to display a list of the first few processes in the class at the bottom of the WLM screen. On the file system section of the main panel of the topas command, press the f key to replace the default report of the system's total file system activity with the busiest file system list. When you press the f key again, you will return to the default screen of total file system activity. The busiest file systems depend on the available space on the screen. In the Volume Group panel (topas-V), you can select the volume group name and press f to display a list of the first few logical volumes that belong to the volume group at the bottom of the LVM panel.

F

Replace the default screen with the full-screen file system screen. This screen provides more detailed information about the file system on the system than the file system section on the home screen. When you press F again, you will return to the default home screen.

G

Activate the Cluster Utilization panel. If the panel is currently active, the G key resets the panel to display the global summary, private, and shared sections. See other cluster utilization dashboard subcommand topics for options specific to this panel.

H

Displays the help screen.

H

Displays the help screen for the local panel, if available.

L

Replace the current screen with a logical partition screen; display LPAR, Micro-Partitioning? that is similar to the content provided by lparstat and mpstat. And concurrent multithreading of metrics.

N

Replace the total network activity report for the system with a list of the busiest interfaces. Press the n key in the network interface subsection. The busiest interfaces displayed will depend on the available space on the screen.

P

Opens and closes the hot process subsection. The number of busiest processes displayed depends on the available space on the screen.

P

Replace the default screen with the full screen process screen. This display provides more detailed information about the processes running on the system than the process section of the home screen. When you press the P key again, it will switch back to the default home screen.

Q

Exit the program.

R

Refresh the screen.

T

Turn the tape screen on or off on the main panel.

T

Displays the full-screen tape screen.

V

Displays the Volume groups panel.

W

Open and close the workload Manager (WLM) class subsection. The number of busiest WLM classes displayed depends on the available space on the screen.

W

Replace the default screen with a full-screen WLM screen. This screen provides more detailed information about the WLM class, the WPAR class, and the processes assigned to the class. When you press the @ key, the WLM class section is replaced with the WPAR subsection. When you press W again, it will switch back to the default home screen.

@

Switch between WLM class metrics and WPAR metrics, that is, monitor WPAR instead of WLM. This key is the @ key. This key is valid for the main panel, processes panel, File system panel, and WLM panel. If you press the @ key on any other panel, the key is ignored. The @ key is restricted within WPAR, that is, it will be ignored within WPAR. The @ key is valid in the following panels:

Main panel

Replace WLM and process subsections with WPAR measures.

Process panel

Replace the default mode of the process panel with the WPAR mode.

File system panel

If you press f, the file system panel will contain the WPAR name. Each WPAR file system metric is displayed in the lower part of the panel.

WLM panel

Replace the WLM subsection with the WPAR subsection.

Arrow key and Tab key

The subsections of the main screen (for example, processors, networks, disks, WLM classes, and full-screen WLM and process screens) can be sorted according to different criteria. Move the cursor over a column and sort by that column. Items are always sorted from the highest to the lowest values. Use the tabbed or arrow keys to move the cursor. Sorting is only valid for 128 disks and 16 network adapters.

~

Displays the nmon screen. This key is the ~ key.

Other cross-partition panel subcommands

When the topas cross-partition panel is active, it accepts the following additional single-character subcommands. At the end of each monitoring interval, the program checks one of the following subcommands and responds to any requested operation.

Project

Description

D

Opens and closes the dedicated partition section.

G

Turn on and off the top global section of the panel to switch between a short list and a detailed list.

R

Forces topas to search for HMC configuration changes when a connection is available. This includes new partitions, processors, or memory allocations found.

S

Open and close the shared partition section.

P

Turns the pool panel section on or off. In the pool panel, the user can select a pool identity and press f to list the shared partitions that belong to the pool.

V

Turns on or off the virtual Iamp O server / client throughput details. You can select a virtual Istroke O server and press the f key to list the VIO clients belonging to that server.

M

Turns the memory pool panel on or off. You can select a memory pool and press f to view the partitions in that pool.

Other cluster utilization panel commands

When the topas cluster utilization panel is active, it accepts the following other single-character subcommands. Each time the monitoring interval passes, the program checks for the existence of one of the following subcommands and responds to any requested action:

Project

Description

D

Opens and closes the dedicated partition section.

G

Turn on and off the top global section of the panel to switch between a short list and a detailed list.

S

Open and close the shared partition section.

Other disk panel (topas-D) subcommands

When the topas disk panel is active, it accepts the following additional single-character subcommands. Each time the monitoring interval passes, the program checks for the existence of one of the following subcommands and responds to any requested action:

Project

Description

D

Turns the Adapters panel on or off.

M

Opens or closes the MPIO panel.

Other adapter panel commands

When the topas adapter panel is active, it accepts the following additional single-character subcommands. Each time the monitoring interval passes, the program checks for the existence of one of the following subcommands and responds to any requested action:

Project

Description

V

Turns the Virtual Adapters panel on or off. Press this key in the adapters panel.

Other logical Partition Panel (topas-L) subcommands

When the topas logical panel is active, it accepts the following additional single-character subcommands. Each time the monitoring interval passes, the program checks for the existence of one of the following subcommands and responds to any requested action:

Project

Description

E

Toggle the Icano memory usage Pool panel.

Other virtual Icano server / client throughput dashboard subcommands

When the topas virtual Isign O server / client throughput panel is active, it accepts the following other single-character subcommands. Each time the monitoring interval passes, the program checks for the existence of one of the following subcommands and responds to any requested action:

Project

Description

D

For the virtual Imax O server that is selected in the Virtual Isign O Server / client Throughput panel, turn on or off the Virtual Isign O server / client disk panel. You can select a server adapter and press f to list the disks and clients that belong to the adapter.

Sample full-screen workload manager class output

The following is an example of a screen generated by the topas-W command:

Topas Monitor for host: ptoolsl3 Interval: 2 Mon Feb 12 06:25:11 2007

WLM-Class (Active) CPU% Mem% Blk-I/O%

System 0 57 0

Shared 0 4 0

Default 0 0 0

Unmanaged 0 14 0

Unclassified 0 38 0

=

DATA TEXT PAGE PGFAULTS

USER PID PPID PRI NI RES RES SPACE TIME CPU% I/O OTH COMMAND

Root 10 108 20 197 9 180 0:24 0.0 0 0 init

Root 1032 0 16 41 3 3374 3 0:00 0.0 00 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 00 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 00 lvmbb

Root 3666 1 108 20 135 23 123 0:00 0.0 00 errdemon

Root 3982 0 108 20 4 3374 4 0:01 0.0 0 0 rtcmd

The following is an example of a screen generated by the topas-W-@ command:

Topas Monitor for host: ptoolsl3 Interval: 2 Mon Feb 12 06:25:11 2007

WPAR CPU% Mem% Blk-I/O%

Neptune001234 01 0

=

DATA TEXT PAGE PGFAULTS

USER PID PPID PRI NI RES RES SPACE TIME CPU% I/O OTH COMMAND

Root 356372 491650 58 41 370 67 370 0:00 0.1 00 topas

Root 262246 188508 24 41 256 21 256 6:27 0.1 0 0 xmtopas

Root 192626 1 60 20 113 17 113 11:17 0.1 0 0 getty

Root 61470 0 16 41 17 0 17 0:31 0.0 0 0 wlmsched

Root 290818 1 58 41 284 67 284 1:54 0.0 0 1 topas

Root 57372 0 37 41 30 0 30 3:39 0.0 0 0 gil

Root 86248 1 60 20 47 0 47 1:04 0.0 0 0 rpc.lock

Root 385224 237728 60 20 254 197 254 0:00 0.0 00 sendmail

Root 131174 176242 60 20 175 79 175 0:03 0.0 0 0 aixmibd

Root 53274 0 36 41 13 0 13 0:05 0.0 0 0 netm

Root 90244 1 60 20 126 2 126 2:35 0.0 0 0 syncd

Root 45078 0 60 41 14 0 14 0:58 0.0 0 0 xmgc

Root 266384 176242 60 20 644 160 644 0:27 0.0 0 0 IBM.CSMA

Root 250004 176242 60 20 617 157 617 0:26 0.0 00 rmcd

Root 184410 176242 60 20 254 197 254 0:14 0.0 0 0 sendmail

Root 151640 0 60 20 13 0 13 0:02 0.0 0 0 rgsr

Root 40980 0 59 41 71 0 71 0:02 0.0 0 0 pilegc

Root 110738 0 60 20 13 0 13 0:01 0.0 0 0 n4bg

Root 180368 1 60 20 98 14 98 0:01 0.0 0 0 cron

Root 10 60 20 158 10 158 0:01 0.0 0 0 init

Example

1. To display up to twenty "hot" disks every five seconds and omit the network interface, WLM class, file system information, and process information, enter the following command:

Topas-i5-n0-p0-w0-f0

two。 To display the five most active processes and up to twenty most active WLM classes (which is the default when the-w flag is omitted), but without network, disk, or file system information, enter the following command:

Topas-p5-n0-d0-f0

3. To run the program with the default options, enter the following command:

Topas

4. To go directly to the process screen, enter the following command:

Topas-P

5. To go directly to the WLM class screen, enter the following command:

Topas-W

6. To go directly to the logical partitioning screen, enter the following command:

Topas-L

7. To go directly to the disk metrics screen, enter the following command:

Topas-D

8. To go directly to the file system screen, enter the following command:

Topas-F

9. To go directly to the WPAR monitoring mode abc, enter the following command:

Topas-@ abc

10. To go directly to topas WPAR mode, enter the following command:

Topas-@

11. To go directly to the LVM screen, enter the following command:

Topas-V

twelve。 To go directly to the tape screen, enter the following command:

Topas-T

13. To go to the shared Ethernet adapter on the VIO Server panel, enter the following command:

Topas-E

14. To go directly to the cluster utilization screen, enter the following command:

Topas-G

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.