lslparutil - list utilization data
Synopsis
Description
Options
Examples
Environment
Bugs
Author
See Also
To list utilization data collected for a managed system:
lslparutil -r {hmc | lpar | pool | procpool | mempool |
sys | all}
-m managed-system
[-d number-of-days] [-h number-of-hours]
[--minutes number-of-minutes]
[--startyear year] [--startmonth month]
[--startday day] [--starthour hour]
[--startminute minute] [--endyear year]
[--endmonth month] [--endday day]
[--endhour hour] [--endminute minute]
[-n number-of-events] [-s sample-rate]
[--filter "filter-data"]
[-F [attribute-names] [--header]] [--help]To list HMC settings for utilization data collection:
lslparutil -r config [-m managed-system]
[-F [attribute-names] [--header]] [--help]
lslparutil lists utilization data collected for a managed-system. This command also lists the Hardware Management Console (HMC) settings for utilization data collection.The HMC collects the following types of utilization data: sampling events, state change events, configuration change events, and Utility Capacity on Demand (CoD) processor usage events.
Sampling events are collected for the managed system, for each partition in the managed system, and for the physical processor pool, each shared processor pool, and the shared memory pool in the managed system. Sampling events are collected at the configured sample rate. They are also collected hourly, daily at midnight, and monthly at midnight on the first day of each month. Sampling events are also collected immediately following the collection of a Utility CoD processor usage event. Sampling events contain information about memory and processor utilization.
State change events are collected for the managed system, for each partition in the managed system, and for the HMC. State change events are collected when a state change occurs for the managed system, a partition, or when the HMC is started or shut down.
Configuration change events are collected for the managed system, for each partition in the managed system, for each shared processor pool in the managed system, for the shared memory pool in the managed system, and for the HMC. Configuration change events are collected when a configuration change affecting memory or processor resources occurs for the managed system, a partition, a shared processor pool, or the shared memory pool. Configuration change events are also collected when the local time is changed on the HMC.
Utility CoD processor usage events are collected for the managed system when a Utility CoD processor minute is used. All Utility CoD processor minutes used during a single minute are grouped into one event.
Hourly sampling events and all state change, configuration change, and Utility CoD processor usage events collected are only saved on the HMC for about 2 months. Daily sampling events are saved for about 2 years, and monthly sampling events are saved for about 10 years.
Sampling events are also saved in a snapshot file. Sampling events that are collected more frequently than hourly are only saved in the snapshot file. All state change, configuration change, and Utility CoD processor usage events collected are also saved in the snapshot file, in addition to being saved with the hourly sampling events. The snapshot file is purged daily at midnight. Events more than 24 hours old are purged. Therefore, the snapshot file will only contain events that were collected during the last 24 to 48 hours.
Utilization data collection for managed systems is not automatically enabled. The chlparutil command can be used to enable utilization data collection and configure the sample rate.
-r The type of system resources for which events are to be listed. Valid values are hmc for HMC, lpar for partitions, pool for physical processor pool, procpool for shared processor pools, mempool for shared memory pool, sys for managed system, all for HMC, partitions, physical processor pool, shared processor pools, shared memory pool, and managed system, and config for HMC configuration settings for utilization data collection. -m The name of the managed system for which the collected utilization data or the HMC configuration settings is to be listed. The name may either be the user-defined name for the managed system, or be in the form tttt-mmm*ssssssss, where tttt is the machine type, mmm is the model, and ssssssss is the serial number of the managed system. The tttt-mmm*ssssssss form must be used if there are multiple managed systems with the same user-defined name, or if the HMC does not currently have a connection to the managed system. If this option is specified when listing HMC configuration settings, then the HMC configuration settings for the managed-system will be listed. Otherwise, the HMC configuration settings for all of the systems currently managed by this HMC, and for all of the systems for which utilization data has been previously collected by this HMC will be listed.
-d The number of days prior to today for which events will be listed. Events that were collected today will be listed, along with events that were collected during the past number-of-days days. This option cannot be specified with the -h, --minutes, --startyear, --startmonth, --startday, --starthour, --startminute, --endyear, --endmonth, --endday, --endhour, or --endminute options.
-h The number of hours prior to the current time for which events will be listed. This option cannot be specified with the -d, --minutes, --startyear, --startmonth, --startday, --starthour, --startminute, --endyear, --endmonth, --endday, --endhour, or --endminute options.
--minutes The number of minutes prior to the current time for which events will be listed. This option cannot be specified with the -d, -h, --startyear, --startmonth, --startday, --starthour, --startminute, --endyear, --endmonth, --endday, --endhour, or --endminute options.
--startyear The starting year for which events will be listed. The default value for this option is 1970. This option cannot be specified with the -d, -h, or --minutes options.
--startmonth The starting month for which events will be listed. Valid values are 1 for January through 12 for December. The default value for this option is 1 (January). This option cannot be specified with the -d, -h, or --minutes options.
--startday The starting day for which events will be listed. Valid values are 1 through 31. The default value for this option is 1. This option cannot be specified with the -d, -h, or --minutes options.
--starthour The starting hour for which events will be listed. Valid values are 0 for midnight through 23 for 11:00 pm. The default value for this option is 0 (midnight). This option cannot be specified with the -d, -h, or --minutes options.
--startminute The starting minute for which events will be listed. Valid values are 0 through 59. The default value for this option is 0. This option cannot be specified with the -d, -h, or --minutes options.
--endyear The ending year for which events will be listed. The default value for this option is now. This option cannot be specified with the -d, -h, or --minutes options.
--endmonth The ending month for which events will be listed. Valid values are 1 for January through 12 for December. The default value for this option is now. This option cannot be specified with the -d, -h, or --minutes options.
--endday The ending day for which events will be listed. Valid values are 1 through 31. The default value for this option is now. This option cannot be specified with the -d, -h, or --minutes options.
--endhour The ending hour for which events will be listed. Valid values are 0 for midnight through 23 for 11:00 pm. The default value for this option is now. This option cannot be specified with the -d, -h, or --minutes options.
--endminute The ending minute for which events will be listed. Valid values are 0 through 59. The default value for this option is now. This option cannot be specified with the -d, -h, or --minutes options.
-n The maximum number of events to be listed, starting with the most recent event. The number specified must be greater than 0. If this option is not specified, and neither are any of the -d, -h, --minutes, --startyear, --startmonth, --startday, --starthour, --startminute, --endyear, --endmonth, --endday, --endhour, or --endminute options, then only the most recent event will be listed.
-s Use this option to specify the sample rate for which utilization data is to be listed. Valid values are h for hourly sampling events and all state change, configuration change, and Utility CoD processor usage events, d for daily sampling events, m for monthly sampling events, and s for the snapshot file. If this option is not specified then events in the snapshot file will be listed.
--filter The filter(s) to apply to the events to be listed. Filters are used to select which events for the specified resource type are to be listed. The filter data consists of filter name/value pairs, which are in comma separated value (CSV) format. The filter data must be enclosed in double quotes.
The format of the filter data is as follows:
"filter-name=value,filter-name=value,..."
Note that certain filters accept a comma separated list of values, as follows:
"" filter-name=value,value,... ",..."
When a list of values is specified, the filter name/value pair must be enclosed in double quotes. Depending on the shell being used, nested double quote characters may need to be preceded by an escape character, which is usually a ’#146; character.
Multiple values can be specified for each filter.
Valid filter names for this command:
lpar_ids
Specify partition ID(s)
lpar_names
Specify partition user-defined name(s)
event_types
Specify one or more of the values
sample, state_change, config_change,
utility_cod_proc_usage
pools
Specify shared processor pool ID(s)
pool_names
Specify shared processor pool
user-defined name(s)
Valid filters with -r hmc:
event_typesValid filters with -r lpar:
event_types, lpar_ids | lpar_names,
pools | pool_namesValid filters with -r pool:
event_typesValid filters with -r procpool:
event_types, pools | pool_namesValid filters with -r mempool:
event_typesValid filters with -r sys:
event_typesValid filters with -r all:
event_types, lpar_ids | lpar_names,
pools | pool_names
The lpar_ids and lpar_names filter will
only apply to partition events, and the pools
and pool_names filter will only apply to
partition events and shared processor pool events-F A delimiter separated list of attribute names for the desired attribute values to be displayed for each event. If no attribute names are specified, then values for all of the attributes for each event will be displayed. When this option is specified, only attribute values will be displayed. No attribute names will be displayed. The attribute values displayed will be separated by the delimiter which was specified with this option.
This option is useful when only attribute values are desired to be displayed, or when the values of only selected attributes are desired to be displayed.
Command attributes:
borrowed_pool_proc_units
The sum of the processing units being borrowed
from powered off partitions with dedicated
processors and the processing units available
to be assigned to partitions, rounded down to
a whole processor.
capped_cycles
The number of capped processing cycles
utilized by this partition since the
managed system was started.
configurable_pool_proc_units
The number of processing units assigned to
all shared processor partitions, rounded up
to a whole processor.
configurable_sys_mem
The amount of configurable system memory
(in megabytes).
configurable_sys_proc_units
The number of configurable system
processing units.
curr_5250_cpw_percent
The 5250 CPW percent assigned to the
partition.
curr_avail_5250_cpw_percent
The 5250 CPW percent available to be
assigned to partitions.
curr_avail_pool_proc_units
The number of processing units available
to be assigned to partitions.
curr_avail_sys_mem
The amount of memory (in megabytes)
available to be assigned to partitions.
curr_avail_sys_proc_units
The number of processing units available
to be assigned to partitions.
curr_io_entitled_mem
The amount of I/O entitled memory (in megabytes)
assigned to the shared memory partition.
curr_max_pool_mem
The maximum size (in megabytes) of the shared
memory pool.
curr_mem
The amount of memory (in megabytes)
assigned to the partition. For shared memory
partitions, this is the amount of logical
memory assigned to the partition.
curr_mem_weight
The current relative memory priority for
the shared memory partition. The smaller the
value, the lower the priority. Possible
values are 0 - 255.
curr_pool_mem
The size (in megabytes) of the shared memory
pool.
curr_proc_mode
The processing mode for the partition.
Possible values are ded or shared.
curr_proc_units
The number of processing units assigned
to the partition.
curr_procs
The number of processors or virtual
processors assigned to the partition.
curr_reserved_pool_proc_units
The number of processing units that are
reserved for temporary use by the
uncapped partitions in the shared
processor pool.
curr_shared_proc_pool_id
The unique integer identifier for the
shared processor pool that the partition
is in.
curr_shared_proc_pool_name
The user-defined name of the shared
processor pool, at the time the event was
collected, that the partition is in.
curr_sharing_mode
The sharing mode of the partition.
Possible values are keep_idle_procs,
share_idle_procs,
share_idle_procs_active,
share_idle_procs_always,
cap, or uncap.
curr_uncap_weight
The current weighted average of
processing priority when in uncapped
sharing mode. The smaller the value, the
lower the weight. Possible values are
0 - 255.
dedup_cycles
The number of processing cycles spent deduplicating
data since Active Memory Deduplication was enabled
for the shared memory pool. The processing cycles
were donated by the paging Virtual I/O Server
partitions assigned to the pool.
dedup_mem
The amount of logical memory (in megabytes) assigned
to the partition that has been deduplicated.
dedup_pool_mem
The amount of memory (in megabytes) in the shared
memory pool that is being used for deduplicated data.
entitled_cycles
The number of processing cycles to which
the partition has been entitled since the
managed system was started. This value is
based on the number of processing units
assigned to the partition, and may be
greater than or smaller than the number
of cycles actually used.
event_type
The type of event. Possible values are
sample, state_change, config_change, or
utility_cod_proc_usage.
idle_cycles
The number of cycles that the partition reported as
idle. If the partition does not have the ability to
report idle cycles, this value is 0.
lpar_curr_io_entitled_mem
The total amount of I/O entitled memory (in
megabytes) assigned to all of the partitions
in the shared memory pool.
lpar_dedup_mem
The total amount of partition logical memory (in
megabytes) that has been deduplicated.
lpar_id
The unique integer identifier for the
partition.
lpar_mapped_io_entitled_mem
The total amount of I/O entitled memory (in
megabytes) currently mapped by all of the
partitions in the shared memory pool.
lpar_name
The user-defined name of the partition
at the time the event was collected.
lpar_run_mem
The total amount of logical memory (in megabytes)
assigned to all of the partitions in the shared
memory pool.
mapped_io_entitled_mem
The amount of I/O entitled memory (in megabytes)
currently mapped by the shared memory partition.
max_pool_proc_units
This number, minus the reserved
processing units in the shared processor
pool, is the maximum number of processing
units that the partitions in the shared
processor pool can use.
mem_dedup
Indicates whether Active Memory Deduplication is
enabled for the shared memory pool. Possible values
are 0 for disabled or 1 for enabled.
mem_mode
The memory mode for the partition.
Possible values are ded or shared.
mem_overage_cooperation
The difference between the shared memory
partition’s assigned memory overcommitment
and its actual overcommitment. A positive
value means the partition is using less memory
than system firmware has requested it to use.
name
The user-defined name of the managed
system.
page_faults
The total number of page faults that have
occurred since the shared memory pool was
created.
page_in_delay
The total page-in delay, in microseconds, spent
waiting for page faults since the shared memory
pool was created.
phys_run_mem
The runtime amount of physical memory (in megabytes)
allocated to the shared memory partition.
prev_time
The time on the HMC when the HMC time was
changed.
proc_cycles_per_second
Processing cycles per second on one physical
processor. This value is static for a particular
managed system.On POWER6 and later servers, this value is set
to a fixed constant of 512,000,000.
resource_type
The type of system resource for which the
event was collected. Possible values are
hmc, lpar, pool, procpool, mempool, or sys.
run_latch_cycles
The number of non-idle cycles used by the partition,
while the run latch was set, since the managed system
was started.POWER6 and POWER7 servers only
run_latch_instructions
The number of non-idle instructions performed by the
partition, while the run latch was set, since the
managed system was started.POWER6 and POWER7 servers only
run_mem_weight
The runtime relative memory priority for
the shared memory partition. The smaller the
value, the lower the priority. Possible
values are 0 - 255.
sample_rate
The rate, in seconds, at which samples
are obtained. This rate can be changed
with the chlparutil command.
shared_cycles_while_active
The number of dedicated processing cycles
shared by this partition while it has
been active since the managed system was
started.
shared_proc_pool_id
The unique integer identifier for the
shared processor pool.
shared_proc_pool_name
The user-defined name of the shared
processor pool at the time the event was
collected.
state
For system events, this is the state of
the managed system at the time the event
was collected. For partition events,
this is the state of the partition at the
time the event was collected.
sys_firmware_mem
The amount of memory (in megabytes) on the managed
system that is being used by system firmware.
sys_firmware_pool_mem
The amount of memory (in megabytes) in the shared
memory pool that is being used by system firmware.
sys_time
The time on the managed system that the
sample was taken.
time
The time on the HMC that the event was
collected.
time_cycles
The number of time cycles since the
managed system was started.
time_shut_down
The time the HMC was shut down or
rebooted.
total_instructions
The number of instructions performed by the partition
since the managed system was started.POWER8 and later servers only
total_instructions_execution_time
The amount of time instruction counts were collected
for the partition since the managed system was
started.POWER8 and later servers only
total_pool_cycles
The total number of processing cycles
available in the physical processor pool
or shared processor pool since the
managed system was started.
type_model_serial_num
The machine type, model, and serial
number of the managed system.
uncapped_cycles
The number of uncapped processing cycles
utilized by this partition since the
managed system was started.
unreported_proc_min
The total number of Utility CoD processor
minutes that have not been reported.
used_proc_min
The number of Utility CoD processor
minutes that were used during the last
minute.
utilized_pool_cycles
The number of processing cycles in the
physical processor pool or shared
processor pool that have been utilized
since the managed system was started.--header Display a header record, which is a delimiter separated list of attribute names for the attribute values that will be displayed. This header record will be the first record displayed. This option is only valid when used with the -F option. --help Display the help text for this command and exit.
List all of the monthly sampling events that were collected for the managed system sys1 for the entire year 2005:lslparutil -r all -m sys1 --startyear 2005 --endyear 2005 --endmonth 12 --endday 31 --endhour 23 -s m
List all of the hourly managed system sampling events that have been collected for the managed system with type, model, and serial number 9406-520*1000101 since midnight today:
lslparutil -r sys -m 9406-520*1000101 -d 0 --filter "event_types=sample" -s h
List all of the events that have been collected for the managed system sys1 since January 1, 2006, and list a maximum of 25 events:
lslparutil -r all -m sys1 --startyear 2006 -n 25 -s h
List the 10 most recent hourly sampling events that were collected for partition p1:
lslparutil -r lpar -m sys1 -n 10 -s h --filter "event_types=sample,lpar_names=p1"
List all of the state change and configuration change events that were collected for the managed system sys1 for the month of June in 2005:
lslparutil -r all -m sys1 --startyear 2005 --startmonth 6 --endyear 2005 --endmonth 6 --endday 30 --endhour 23 -s h --filter ""event_types=state_change,config_change""
List all of the Utility CoD processor usage events that were collected for the managed system sys1 for the month of May in 2007:
lslparutil -r all -m sys1 --startyear 2007 --startmonth 5 --endyear 2007 --endmonth 5 --endday 31 --endhour 23 -s h --filter "event_types=utility_cod_proc_usage"
List all of the HMC events that have been collected for managed system sys1 since February 1, 2006:
lslparutil -r hmc -m sys1 --startyear 2006 --startmonth 2 -s h
List the configuration settings for utilization data collection on this HMC for all managed systems:
lslparutil -r config
To calculate the default shared processor pool utilization in percent over a twelve hour time period:
lslparutil -m sys1 -r procpool --startyear 2006
--startmonth 2 --startday 23 --starthour 0 --endyear 2006
--endmonth 2 --endday 23 --endhour 12 --filter
"event_types=sample,pool_names=DefaultPool"
-F time,total_pool_cycles,utilized_pool_cycles02/23/2006 12:00:01,134967149091025,467439053292
02/23/2006 11:00:02,134963299532241,467428119008
02/23/2006 10:00:01,134959313365305,467419269942
02/23/2006 09:00:01,134954622214624,467403199531
02/23/2006 08:00:02,134942086330068,467368397739
02/23/2006 07:00:01,134929553859752,467333227651
02/23/2006 06:00:01,134917026289150,467295577359
02/23/2006 05:00:02,134904482088726,467258616569
02/23/2006 04:00:01,134891946956456,467223704573
02/23/2006 03:00:01,134879415157938,467188374373
02/23/2006 02:00:01,134866883128692,467152556956
02/23/2006 01:00:02,134854347365860,467116506907
02/23/2006 00:00:03,134841811733640,467081011935Pool utilization =
(utilized_pool_cycles / total_pool_cycles) * 100
Pool utilization = ((467439053292 - 467081011935) /
(134967149091025 - 134841811733640)) * 100
Pool utilization = 0.29%To calculate the processor utilization for the shared processor partition with ID 7 over the last minute (the sampling rate is 1 minute):
lslparutil -m sys1 -r lpar --filter "lpar_ids=7", -n 2 -F time,lpar_id,
capped_cycles,uncapped_cycles,entitled_cycles,time_cycles03/29/2012 18:09:30,7,55844419525044,0,106332165468039,
1234978811234159
03/29/2012 18:08:30,7,55839107527864,0,106316742434446,
1234947965168143Processor utilization % =
((capped_cycles + uncapped_cycles) / entitled_cycles) * 100
Processor utilization % =
(((55844419525044 - 55839107527864) + (0 - 0)) /
(106332165468039 - 106316742434446)) * 100
Processor utilization % = 34.44%Processor units utilized =
(capped_cycles + uncapped_cycles) / time_cycles
Processor units utilized =
((55844419525044 - 55839107527864) + (0 - 0)) /
(1234978811234159 - 1234947965168143)
Processor units utilized = 0.17To calculate the processor utilization for the dedicated processor partition with ID 8 over the last minute (the sampling rate is 1 minute):
lslparutil -m sys1 -r lpar --filter "lpar_ids=8", -n 2
-F time,lpar_id,capped_cycles,idle_cycles,time_cycles03/29/2012 18:09:30,8,10650723610566,27015925200832,1234978811234159
03/29/2012 18:08:30,8,10619877543196,26994400296144,1234947965168143Processor utilization % =
((capped_cycles - idle_cycles) / capped_cycles) * 100
Processor utilization % = (((10650723610566 - 10619877543196) -
(27015925200832 - 26994400296144)) /
(10650723610566 - 10619877543196)) * 100
Processor utilization % = 30.22%Processor units utilized =
(capped_cycles - idle_cycles) / time_cycles
Processor units utilized = ((10650723610566 - 10619877543196) -
(27015925200832 - 26994400296144)) /
(1234978811234159 - 1234947965168143)
Processor units utilized = 0.30
None
None
IBM Austin
chlparutil, rmlparutil
| Linux | LSLPARUTIL (1) | "June 2014" |