CPC LPAR Details History attributes

This attribute group provides historical information about CPU usage for the highest CPU consumer in a sysplex or CPC. This group also provides partition management statistics for the LPAR identified in the data row.

Absolute Capacity Limit (RMF metric) The absolute physical hardware capacity limit of the LPAR; this is the number of physical CPUs an LPAR is limited to, in hundredths of a CPU. Set by the operator in the logical partition controls on the Hardware Management Console. "Unavailable" is displayed when the RMF does not return a value for this attribute. A valid value is a number to two decimal places accuracy. Values appear on the workspace as follows:

Actual MSUs Consumed (RMF metric) The actual millions of unweighted CPU service units per hour consumption. These values are only provided for general purpose processors. A valid value is an integer. Values appear on the workspace as follows:

Average Logical CPs (RMF metric) The average number of logical processors which were online during the report interval. A valid value is a number to one decimal places accuracy. Values appear on the workspace as follows:

Capping Option (RMF metric) The capping option of the partition (Yes, No, Mixed or Unavailable). This attribute indicates whether the operator has set "Initial Capping ON" or an absolute physical hardware capacity limit in numbers of CPUs in the logical partition controls for the partition. Mixed is set by RMF, if either a non-IBM processor belongs to this partition which is not managed by the logical partition controls or if the capping status is currently changing. A valid value is an integer. Values appear on the workspace as follows:

CPC Serial Number (RMF metric) The machine serial number (last six hexadecimal digits). A valid value is a string of up to 6 characters.

Current Weight (RMF metric) The current weighting of the shared processor resources for the partition. A valid value is an integer. Values appear on the workspace as follows:

This metric is available for general purpose processors (CP) as well as for special purpose processors (for example, ICF). With LPAR weight management, you give each logical partition an initial LPAR weight, along with an optional minimum and maximum weight if desired.

Workload Manager (WLM) then dynamically balances these weights to best meet the goals of the work in the partitions, with no human intervention. Note that the total weight of the cluster as a whole will remain constant, so LPARs outside the cluster are unaffected.

Defined Capacity (RMF metric) Indicates whether or not the LPAR has defined capacity, A valid value is an integer. Values appear on the workspace as follows:

Defined MSU Capacity (RMF metric) The defined Capacity of the LPAR in millions of unweighted CPU service units per hour (MSU/Hr). If no defined capacity is specified this is the Entitled Capacity. These values are only provided for general purpose processors. A valid value is an integer. Values appear on the workspace as follows:

Effective Logical CP Percent (RMF metric) The effective logical CP percentage for the LPAR. A valid value is a number to one decimal places accuracy. The range for this value is from 0 to 100.0. Values appear on the workspace as follows:

This data is based on the total online time of all logical processors and does not include LPAR management time.

This metric is available for partitions running general purpose processors (CP) as well as for partitions running special purpose processors (for example, ICF).

Effective Physical CP Percent (RMF metric) The effective physical CP percentage for the LPAR. The effective utilization of the physical processor resource by the partition. A valid value is a number to one decimal places accuracy. The range for this value is from 0 to 100.0. Values appear on the workspace as follows:

(Effective Dispatch Time / # Physical Processors * Range Time) * 100

High Share LPs (RMF metric) The number of logical processors in the LPAR with a high entitlement (100% share) of a physical processor. A valid value is an integer. Values appear on the workspace as follows:

Initial Capping Option (RMF metric) The initial capping option of the partition (Yes, No, Mixed or Unavailable). This field indicates whether the operator has set 'Initial Capping ON' in the logical partition controls for the partition. Mixed is set by RMF, if either a non-IBM processor belongs to this partition which is not managed by the logical partition controls or if the Initial Capping status is currently changing. A valid value is an integer. Values appear on the workspace as follows:

Initial Weight (RMF metric) The defined initial weighting of the shared processor resources. A valid value is an integer. Values appear on the workspace as follows:

With LPAR weight management, you give each logical partition an initial LPAR weight, along with an optional minimum and maximum weight if desired. Workload Manager (WLM) then dynamically balances these weights to best meet the goals of the work in the partitions, with no human intervention. Note that the total weight of the cluster as a whole will remain constant, so LPARs outside the cluster are unaffected.

Low Share LPs (RMF metric) The number of logical processors in the LPAR with a low entitlement (0% share or very close to it) of a physical processor. A valid value is an integer. Values appear on the workspace as follows:

LP %Share of Physical (RMF metric) The percentage share of each Hiperdispatch Medium Logical Processor. The percentage of the physical processor that a logical processor of the LPAR is entitled to use.

If HiperDispatch is enabled, this is the percentage of logical processors with medium entitlement of a physical processor.

Without HiperDispatch, all logical processors have the same share because the LPAR weight is evenly distributed.

Blue highlighting of data values indicates that the LPAR is in HiperDispatch mode. A valid value is a number to one decimal places accuracy. Values appear on the workspace as follows:

LPAR 4 Hour Rolling Average (RMF metric) The average millions of unweighted CPU service units per hour (MSU/Hr) consumed during the last four hours in the LPAR. A valid value is an integer. Values appear on the workspace as follows:

LPAR Capping Percentage (RMF metric) The percentage of time when WLM capped the LPAR because the four-hour rolling average MSU value exceeded the defined capacity limit. A valid value is a number to one decimal places accuracy. The range for this value is from 0 to 100.0. Values appear on the workspace as follows:

LPAR Cluster Name (RMF metric) The name of the LPAR Cluster this LPAR belongs to. All partitions that have the same cluster name are grouped together. A valid value is a string of up to 8 characters.

LPAR Configured Storage MB (RMF metric) The central storage currently online to this LPAR in Megabytes. A valid value is an integer. Values appear on the workspace as follows:

LPAR CP Type (RMF metric) The type of processor associated with the current row's statistics. A valid value is an integer. Values appear on the workspace as follows:

LPAR Group Active 4 Hours (RMF metric) Indicates whether or not the LPAR Group has been active for 4 hours or more. A valid value is an integer. Values appear on the workspace as follows:

LPAR Group Average Unused MSUs (RMF metric) The LPAR group average 4 hour unused MSUs/Hour. A valid value is an integer. Values appear on the workspace as follows:

LPAR Group Capacity Limit (RMF metric) LPAR Group Capacity MSU Limit. Capacity limit (in MSUs) defined for the partition's capacity group. A valid value is an integer. Values appear on the workspace as follows:

LPAR Group Maximum Entitlement (RMF metric) The maximum entitled MSU/Hr of the LPAR Capacity Group this LPAR belongs to. The maximum MSU share a partition can get if all other partitions in the capacity group are running without workload. A valid value is an integer. Values appear on the workspace as follows:

LPAR Group Minimum Entitlement (RMF metric) The minimum entitled MSU/Hr of the LPAR Capacity Group this LPAR belongs to. The guaranteed MSU share this partition gets if necessary (even if all other partitions within the capacity group are running high workload). A valid value is an integer. Values appear on the workspace as follows:

LPAR Group Name (RMF metric) The name of the partition's capacity group. A valid value is a string of up to 8 characters. If the partition does not belong to a capacity group, a blank is displayed.

LPAR Group Time Until Capping (RMF metric) The projected time until this LPAR's LPAR Group will be subject to capping. A valid value is an integer. Values appear on the workspace as follows:

LPAR Name (RMF metric) The partition name. A valid value is a string of up to 8 characters.

Note:

  1. Partitions identified by the name PHYSICAL are not configured partitions. Data reported in these lines includes the time during which a physical CPU was busy, but the time could not be attributed to a specific logical partition.
  2. Starting with z9 processors, IFLs (Integrated Facility for Linux) and zAAPs are reported separately and no longer as ICFs (Internal Coupling Facility).

LPAR Time Until Capping (RMF metric) The projected time until this LPAR will be subject to capping. A valid value is an integer. Values appear on the workspace as follows:

Managed System z/OS operating system in your enterprise that an OMEGAMON for z/OS agent is monitoring. A valid value is a string of up to 32 characters.

Maximum 4 Hour Rolling Average (RMF metric) The maximum value of consumed MSUs during the last 4 hours (retrieved from 48 sample intervals of five minutes). This value can be greater than the defined capacity. A valid value is an integer. Values appear on the workspace as follows:

Maximum Weight (RMF metric) The maximum weighting of the shared processor resources for the partition. A valid value is an integer. Values appear on the workspace as follows:

A value of zero indicates that the partition is under control of LPAR CPU management, but no maximum has been specified. With LPAR weight management, you give each logical partition an initial LPAR weight, along with an optional minimum and maximum weight if desired.

Workload Manager (WLM) then dynamically balances these weights to best meet the goals of the work in the partitions, with no human intervention. Note that the total weight of the cluster as a whole will remain constant, so LPARs outside the cluster are unaffected.

Medium Share LPs (RMF metric) The number of logical processors of the LPAR with a medium entitlement of a physical processor. A valid value is an integer. Values appear on the workspace as follows:

For the percentage, refer to the LP %Share of Physical value.

Minimum Weight (RMF metric) The minimum weighting of the shared processor resources. A valid value is an integer. Values appear on the workspace as follows:

A value of zero indicates that the partition is under control of LPAR CPU management, but no minimum has been specified. With LPAR weight management, you give each logical partition an initial LPAR weight, along with an optional minimum and maximum weight if desired.

Workload Manager (WLM) then dynamically balances these weights to best meet the goals of the work in the partitions, with no human intervention. Note that the total weight of the cluster as a whole will remain constant, so LPARs outside the cluster are unaffected.

Physical CP Overhead Percent (RMF metric) LPAR physical overhead percentage. A valid value is a number to one decimal places accuracy. The range for this value is from 0 to 100.0. Values appear on the workspace as follows:

The average LPAR management time percentage.

((Total Dispatch Time - Effective Dispatch Time)/(# Physical Processors * Range Time)) * 100

The calculation for the PHYSICAL partition is:

(Time PHYSICAL /(# Physical Processors * Range Time)) * 100

Time PHYSICAL is the time that could not be attributed to a specific logical partition but was used by PR/SM to control the physical processor (LPAR management time).

Recording End Time Time at the end of the RMF reporting interval for the current row of near-term historical data. This time is displayed in the following format:

mm/dd/yy hh:mm:ss (Tivoli Enterprise Portal) or yy/mm/dd hh:mm:ss (3270)

Where:

This time is stored as a string no longer than 16 characters in the format CYYMMDDHHMMSSmmm (as in 1020315064501000 for 03/15/02 06:45:01) where:

Recording Start Time Time at the start of the RMF reporting interval for the current row of near-term historical data. This time is displayed in the following format:

mm/dd/yy hh:mm:ss (Tivoli Enterprise Portal) or yy/mm/dd hh:mm:ss (3270)

Where:

This time is stored as a string no longer than 16 characters in the format CYYMMDDHHMMSSmmm (as in 1020315064501000 for 03/15/02 06:45:01) where:

RMF Status Status indications for availability of RMF and lock data. A valid value is an integer. Values appear on the workspace as follows:

Sample Period Sample period in seconds. A valid value is an integer.

Total Logical CP Percent (RMF metric) The total utilization of the logical processors by the partition. A valid value is a number to one decimal places accuracy. The range for this value is from 0 to 100.0. Values appear on the workspace as follows:

This data is based on the total online time of all logical processors and includes LPAR management time.

This metric is available for partitions running general purpose processors (CP) as well as for partitions running special purpose processors (for example, ICF).

Total Physical CP Percent (RMF metric) The total utilization of the physical processor resource by the partition. A valid value is a number to one decimal places accuracy. The range for this value is from 0 to 100.0. Values appear on the workspace as follows:

(Total Dispatch Time /# Physical Processors * Range Time) * 100

The Total Dispatch Time for the PHYSICAL partition includes the time during which a physical CPU was busy, but the time could not be attributed to a specific logical partition. This time includes the time PR/SM was controlling the physical processor (LPAR management time), as well as any other time the processor was busy for any reason such as managing coupling facility traffic.