Active workload target distribution factors - WLMAWAOR

The Target regions in an active workload (WLMAWAOR) views display information about all target regions that are associated with a workload that is within the CICSplex identified as the context.

Supplied views

To access from the main menu, click:

Active workload views > Active workload target distribution factors

Table 1. Views in the supplied Target region in active workload (WLMAWAOR) view set
View Notes
Target region in active workload

EYUSTARTWLMAWAOR.ACTIVATE

Cause the specified target region to be recognized so that it can participate in workload management.
Target region in active workload

EYUSTARTWLMAWAOR.DETAILED

Detailed information about a selected target region.
Target region in active workload

EYUSTARTWLMAWAOR.QUIESCE

Cause the specified target region to be removed from workload management.

This means that no further transactions are routed to the target region unless they share an affinity with earlier transactions. For example, if you are engaged in a pseudoconversation when a target region begins quiescing, the transactions that constitute the pseudoconversation continue to be routed to the same target region until the end of the affinity lifetime. When the affinity is no longer active, the target region is fully quiesced and subsequent transactions are routed to a different target region in the target scope.

Target region in active workload

EYUSTARTWLMAWAOR.TABULAR

Tabular information about target regions that are associated with a workload that is within the CICSplex identified as the context.
Target region in active workload

EYUSTARTWLMAWAOR.TABULAR2

Tabular information about target regions that are associated with a workload that is within the CICSplex identified as the context.

Actions

Table 2. Actions available for WLMAWAOR views
Action Description
ACTIVATE Cause the specified target region to be recognized so that it can participate in workload management.
QUIESCE Cause the specified target region to be removed from workload management.

This means that no further transactions are routed to the target region unless they share an affinity with earlier transactions. For example, if you are engaged in a pseudoconversation when a target region begins quiescing, the transactions that constitute the pseudoconversation continue to be routed to the same target region until the end of the affinity lifetime. When the affinity is no longer active, the target region is fully quiesced and subsequent transactions are routed to a different target region in the target scope.

Fields

Table 3. Fields in WLMAWAOR views
Field Attribute name Description
Target region name AOR The name of an active CICS system, acting as a target region, to which the workload is associated.
Coupling Facility update count CFUPDCNT This is the number of update calls that the RS Server has made to the z/OS Coupling Facility for this target region.

This value is reset to zero whenever the RS Server or the CICS region is started.

Target region contact status CON_STATUS The status of the connection between the target region and the CMAS. Values are:
  • LOSTCON - The connection between the target region and the CMASs has been lost.
  • N_A - The connection between the target region and the CMASes managing the workload is available.
CICSPlex SM version CPSMVER This is the CICSPlex SM version of the target region.
RTA event active EVENTS This is the indicator of an outstanding RTA event for the current region. It is associated with either the workload specification or if the Transaction name (TRANNAME) parameter is specified and does not have a value of '*', then the TRANGRP in the current workload which manages that specific transaction. This value may vary for each workload that targets the current region, and will cause corresponding variations of the routing weight for that region.
Region is dumping HLTHDUMP This indicates whether the routing target is currently issuing a transaction or system dump.
Region is at MAXTASKS HLTHMAXT This indicates whether the routing target is currently running at a task rate that has reached its maximum tasks threshold.
MAS non responsive HLTHNRM This field indicates the routing target is non responsive at the current time.
Short on storage HLTHSOS This field indicates the SOS status of this routing target at the current time. Note that a CICS region can change from being in and out of SOS status very rapidly.
Region has stalled HLTHSTALL This indicates whether the routing target is in a stall condition. If the region is stalled, use the Active task views to determine which tasks are suspended, and determine why. Note: Stall here means a task has been waiting for the same resource for a specific time.
Maximum task count for region MAXTASKS This is the defined maximum number of active tasks that may concurrently execute in this routing target.
Optimization status OWSTATE This field reports the status of the current target region for the optimized workload routing function.

The possible values are:

  • ACTIVE
    • This target region is executing in an optimized workload state.
  • INACTIVE
    • This region is capable of executing in an optimized workload state, however, it is not currently optimized for one or more of the following reasons:
      • The region has no connection to an RS server
      • The region is connected to an RS server, however, the server has not been able to connect to the z/OS coupling facility
      • The optimised workload routing function is DISABLED for the current target region
      • The RS server update frequency value for this region is 0, which means that the optimization capabilities for this region are not enabled
  • N_A
    • The target region is not configured to provide optimized workload management. Only non optimized WLM routing decisions can be made.

If the CMAS that your WUI is using is not connected to a Region Status (RS) server, the optimization status for remote regions will not be updated and defaults to INACTIVE. Where this situation occurs, either connect all reporting CMASs to the RS server or use the Active target regions view to check the optimization state.

RS server read interval READRS Within a workload running in optimized mode, this value specifies the minimum interval between refreshes of a target region status from a CICS CFDT Server. These refresh requests will be issued by a routing region that is evaluating this region as a possible target for a dynamic routing request.

The value range is from 0 to 2000, and represents units of milliseconds:

  • When the Optimization status is not set to Active, then this value will be set to 0, and WLM will ignore it. When the Optimization status is set to Active, then a value of 0 means that a routing region will request a status update of a target region on every occasion that it examines this region's status.
  • Values between 1 and 2000 specify the minimum time interval that must expire before the status of this region can be refreshed.

    A low interval value means that the CFDT Server will be polled more often for a status update, than for a higher value. For workloads in QUEUE mode, this will result in a task load more evenly distributed across the CICS regions in the workload target scope (assuming all other health and link factors are equal) . However, the utilization of the RS server will be correspondingly increased, which may consequently result in a higher utilization of your z/OS Coupling Facility.

A value of 1000 represents a one second interval. The default value is 200 milliseconds

This value may be modified for this specific region using the MASs known to CICSplex views, or the CICS system definition views.

WLM routing weight for region ROUTEWGHT This is the calculated routing weight for the current routing target according to a specified transaction name as specified in the Transaction name (TRANNAME) parameter. If the transaction name has a value of '*', the routing weight is that of the default route suggested by the workload specification. This value will be combined with the link weight from a Router to determine an overall normalized routing weight for this CICS region. This value will be compared with the same value for other potential routing targets. Assuming no affinities are outstanding, then the target with the lowest routing weight will be chosen as the actual route target.

If the workload associated with this target region specifies that RTA MRM events should influence a route decision for certain transaction codes, then different transaction names will result in different routing weights because of the effects of outstanding RTA events.

Note: 2147483647 is a special value indicating that the target is quiesced.

Generally speaking, the Link weight is used to prefer faster connections over slower ones. Weighting is as follows;

  • MRO is considered faster than XCF (cross coupling-facility) connections.
  • XCF is considered faster than IPIC.
  • IPIC is considered faster than VTAM.
  • VTAM is considered faster than indirect connections.
Routing Load ROUTINGLOAD This is the routing load used by WLM route decision making for the target region. Note that this will not match the real CICS task count at the time of this query.

For an ACTIVE System Optimized WLM routing region this will be the approximate task load in the target region, taking into account 'Task Count Increment Value', 'RS Server Bottom Tier' and 'RS Server Top Tier'. The accuracy of this value is also affected by the RS Read Interval and RS Update frequency settings.

For target regions that are not connected to a RS Server, and are therefore in a non-optimized WLM state, this value will appear to be less accurate, when compared to the actual task count. This is because this is the normalized routing load for this region. It should be noted that when running in a non-optimized WLM state, this value can exceed the MAX TASK value, without the MAX TASK flag being set on, and is correct behaviour as the MAX TASK indicator is related to the target region's health status taken from actual task counts.

It is possible that a region could be the target for multiple workloads and show different routing weight values. This is not abnormal behaviour - it is a symptom of the synchronization latencies within workload manager. Workload optimization will nullify this effect. The workload manager runtime processes will continue to execute normally in spite of this disparity in the load values. To suppress these observed load differences, try to ensure that all MASes and CMASes that participate in the workload have a connectable region status server available.

Reporting CMAS RPTCMAS The name of the CMAS that provided the information in this record.

If the GET operation optional parameter EXPLODE is not specified or is specified as NO, then this attribute will be set to blanks and only one record will be returned for the target region.

If the GET operation optional parameter EXPLODE is specified as YES, then this attribute will be set with the name of the CMAS that provided the information in this record. A separate record will be returned for the target region from each CMAS that manages the workload associated with the target region.

When the IBM distributed WUI menus and views are used to request the WLMAWAOR tabular view, the EXPLODE parameter will be set to YES.

RS server pool name RSPOOLID For sysplex optimized workloads, region status data is maintained within a coupling facility data table. That table will be contained within a CFDT pool identified by this attribute. For an optimized workload to activate, there must be an active Region Status server which manages data for the pool name identified here.

If your CICSplex identifiers are not unique within your sysplex, you must ensure that the RS server pool names are unique. If your Sysplex comprises unique CICSplex identifiers, then they may all share the same RS server pool.

This attribute is defined in the parent CICSplex definition of the current target region. The default name is DFHRSTAT. You may choose to employ an existing CFDT pool for containing your CICSplex data tables. If you do, be aware that the throughput of your optimized workloads may be impeded by any user application activity to the specified pool name. Likewise, any application throughput to the pool may be impacted by sysplex optimized workloads. It is recommended that a discrete RS server and pool name is defined for the optimized workload function.

Target region status STATUS The current status of the target region associated with the workload, as one of the following:
  • ACTIVE - The target is available to participate in workload management.
  • QUIESCEING - The target is being quiesced. No new transactions are routed to this target. Transactions currently being routed to this target are routed to another target, unless an affinity relation exists. If there is an affinity relation, transactions continue to be routed to this target until the affinity lifetime expires. Note that if a target has an affinity lifetime of PERMANENT, it will remain in a QUIESCING state indefinitely.
  • QUIESCED - The target is not available to participate in workload management.

Input Values: ACTIVE | QUIESCE

Task count increment value TASKINC This is the translation of the RS Server update frequency value to an actual task count value for the current target region. Each time the current task count reaches a boundary that is a factor of this value, then an update is sent to the RS server.
Task load percentage TASKLOAD This is the normalized task load for this target region. It is calculated from dividing the current task routing count by the MAXTASKS value for the region, and multiplying by 100.
RS Server update frequency UPDATERS Within a workload running in optimized mode, this value specifies the frequency with which the CICS CFDT (RS) server will be called to modify the value of the task load within this target region.

The value range is from 0 to 25:

  • When the Optimization status is not set to Active, then this value will be set to 0, and WLM will ignore it. When the Optimization status is set to Active, then a value of 0 means that the RS Server is not notified of any task load count changes, which disables the optimized workload function for this target region.
  • Values between 1 and 25 are applied as an arithmetic percentage to a region's MAXTASKS setting. The resultant task count value is used as a numeric threshold to drive an update call to the RS Server.

For example , with a MAXTASKS setting of 120, and with this attribute set to 20, the RS Server will be called to update the WLM load count when the regions task count changes between:

  • 23 and 24 tasks - (20%),
  • 47 and 48 tasks - (40%),
  • 71 and 72 tasks - (60%),
  • 95 and 96 tasks - (80%),
  • 119 and 120 tasks - (100%).

The RS Server would be updated when the task load increments or decrements across these boundaries.

If a value is specified that is at the lower end of the 1-25 scale, then that will cause an increase in the frequency of updates to the RS Server across the task load range. For workloads in QUEUE mode, this will result in a task load more evenly distributed across the CICS regions in the workload target scope (assuming all other health and link factors are equal) . However, the utilization of the RS server will be correspondingly increased, which may consequently result in a higher utilization of your z/OS Coupling Facility.

This value may be modified for this specific region using the MASs known to CICSplex views, or the CICS system definition views.

z/OS WLM health indicator WLMHLTH The percentage value that represents the z/OS WLM health open state for the region. A value of zero means that z/OS WLM will consider this region to be in an unhealthy state, and its routing weight will be correspondingly increased, making it a less desirable target region compared with those in a target scope whose values are greater than zero. A value of 100 means either that a region has now achieved a fully open z/OS WLM health open state, or that the region is running with WLMHEALTH=OFF in the SIT and a fully opened z/OS WLM health open state is assumed by default. Likewise a value of 100 is returned for a region at a lower CICS level than CICS TS 5.4 for which a fully opened z/OS WLM heath open state is assumed by default.
Task load queue mode WLMQMODE This attribute specifies how the queued task load of a target CICS region is to be evaluated:
  • MAXTASK - specifies that both active and MAXTASK queued tasks are to be included in the task load evaluation for the region.
  • ALL - specifies that the task load evaluation for the region will include active tasks, tasks queued for the MAXTASK limit and tasks that are queued because of a TRANCLASS limit.
  • N_a - specifies that the current region is not at a high enough level to support the use of this attribute. You will not be able to change it, and the WLM route decision process will ignore it.

The default value is ALL. This value is specified within the CICSPlex SM CICS system definition object. If you want this value to be changed, you must modify the system definition for the current region and then restart it. You may change the value in an active CICS region by using the CICSPlex SM MAS object. When the region restarts it will revert to the value specified in the system definition.

Task load health threshold WLMTHRSH This specifies a percentage threshold of the current region's task load, which is calculated by dividing the current task count by the maximum task count. When the load for a target region reaches this threshold, then WLM considers the region to be relatively unhealthy. This will cause higher link weights to be applied to the WLM routing algorithm when evaluating this region.

When a target scope covers local and remote regions relative to the router, then WLM will favour local targets over remote ones. The effect of this attribute is that when this load threshold in the local regions is achieved, then WLM will start to favour remote targets instead. When all targets in the scope achieve this load threshold, then WLM will revert to favouring local targets again.

The value range is from 1 to 100, and the default value is 60. This value is specified within the CICSPlex SM CICS System Definition object. If you want this value to be changed, you must modify the system definition for the current region and then restart it. You may change the value in an active CICS region by using the CICSPlex SM MAS object. When the region restarts it will revert to the value specified in the system definition.

If this value is set to 0, then the current region is not at a high enough level to support the use of this attribute. You will not be able to change it, and the WLM route decision process will ignore it.

Note: this value is nullified when applied to the routing factor of link neutral dynamic routing requests. This is because the link weight itself is ignored for the LNQUEUE and LNGOAL algorithms.

Workload name WORKLOAD The name of the workload specification.
System ID of workload owner WRKLOWNER The 4-character CICS system ID of the CMAS that created the workload.


eyuaae4.html | Timestamp icon Last updated: Sunday, 4 June 2017