Execution Policy
Description
This container executes the functors it contains under a specified processor and parallelization policy, temporarily overriding the policy otherwise in effect for as long as its content runs.
Inputs
None.
Optional Inputs
| Name | Type | Description | Default Value |
|---|---|---|---|
| Maximum Granularity | Non Negative Integer Value Type | Maximum granularity used to split a job into parallel tasks. Set to 0 to use the default granularity. | 0 |
| Force Specified Granularity | Boolean Value Type | Always use the granularity specified in Maximum Granularity, even when the current granularity is already within that limit. | No |
| Computing Target | Enum Type | How the contained functors should be run: Default suggests the use of accelerator computing devices when available, while still allowing regular processors to be used; Processors Only forces the use of regular processors; Accelerator Devices Only forces the use of accelerator computing devices, and raises an error during execution if none are available. | None |
| Sequence Input | None Type | Ensures that the functor connected to this port finishes executing before this container starts. | None |
Outputs
| Name | Type | Description |
|---|---|---|
| Sequence Output | None Type | Connect this port to another container’s Sequence Input to make this container finish executing before that other container starts. |
Group
Notes
See Parallel Execution for how granularity and accelerator devices are used.
This functor was previously known as “ProcessorPolicy”.
The policy in effect before this container ran is always restored once its content finishes executing, whether or not an error occurs.
Internal Name
ExecutionPolicy