ACTIVITYTOTALRUNTIMEINALLSC threshold

The ACTIVITYTOTALRUNTIMEINALLSC threshold specifies the maximum run time for an activity in a service subclass.

Type
Activity
Definition domain
Service subclass
Enforcement scope
Database
Tracked work
Recognized coordinator and nested activities (see Activities)
Queuing
No
Unit
Time
Predictive or reactive
Reactive
The ACTIVITYTOTALRUNTIMEINALLSC threshold is identical to the ACTIVITYTOTALRUNTIME threshold with the following exceptions:
  • The ACTIVITYTOTALRUNTIMEINALLSC threshold can be defined only on the service subclass domain.
  • The ACTIVITYTOTALRUNTIMEINALLSC threshold supports the REMAP action.

When an activity violates any threshold with a REMAP action and is moved to a new service subclass, any ACTIVITYTOTALRUNTIMEINALLSC thresholds that are currently applied to the activity from the original service subclass are removed. Any ACTIVITYTOTALRUNTIMEINALLSC thresholds from the original service subclass are no longer in effect for the activity and replaced by any ACTIVITYTOTALRUNTIMEINALLSC thresholds that are defined on the new service subclass. If no ACTIVITYTOTALRUNTIMEINALLSC thresholds are defined on the new service subclass, then the activity is no longer governed by an ACTIVITYTOTALRUNTIMEINALLSC threshold.

The ACTIVITYTOTALRUNTIMEINALLSC threshold always measures runtime relative to the start of activity execution. This behavior is different from other remapping thresholds, such as CPUTIMEINSC, which tracks resource consumption only for the portion of time that the activity runs in that service class. This difference is the reason a new suffix INALLSC is introduced for the RUNTIME remapping threshold instead of using the existing “INSC” suffix. These labels are based on the following convention:
INSC
The threshold measures resource consumption or time in a service class. In other words, resource or time is measured in the current service class where the activity is running.
INALLSC
The threshold measures resource consumption or time in all service classes. In other words, resource or time is measured across all service classes where the activity is running, or ran.
An example follows, based on an environment with these thresholds:
  • A CPUTIMEINSC threshold in service class A remaps an activity to service class B after 20 seconds of CPU is used.
  • A CPUTIMEINSC threshold in service class B terminates activities that use more than 10 seconds of CPU time.
  • An ACTIVITYTOTALRUNTIMEINALLSC threshold in service class B terminates activities that run for more than 30 seconds.
Suppose an activity starts at time T in service class A. After the activity uses 20 seconds of CPU time, it moves to service class B. When the activity starts running in service class B, the activity terminates if either of the following conditions are met:
  • The activity uses 10 more seconds of CPU
  • The absolute run time of the activity exceeds T + 30 seconds.

The REMAP behavior for the ACTIVITYTOTALRUNTIMEINALLSC threshold also works differently from REMAP on the in service class thresholds in a partitioned database environment. When an activity is remapped by the ACTIVITYTOTALRUNTIMEINALLSC threshold in a partitioned database environment, the activity is remapped on all database partitions where it is running. In contrast, the in service class thresholds, such as CPUTIMEINSC, are evaluated per partition and remap the activity only on the partition (or partitions) where the thresholds are violated.