This configuration parameter is available only if you have configured HPE Media Server to use a GPU. For information about using a GPU to accelerate analysis, refer to the HPE Media Server Administration Guide.
The maximum amount of time to wait before sending a batch of video frames to the GPU for analysis. After a video frame has been waiting in the queue for this amount of time, video frames are sent to the GPU even if the number of frames is less than that specified by GPUNumParallel.
To maximize throughput and use the GPU most efficiently, configure a long GPUBatchingDuration
so that HPE Media Server waits until there is a full batch of video frames to analyze. If you are processing low volumes of frames or require the analysis results as rapidly as possible, you can reduce the duration.
Type: |
![]() The value of this parameter is a time duration. Specify the duration in the following format: 1days 2hours 3minutes 4seconds 500milliseconds 600microseconds You can omit any of the individual durations when they are not required: 4seconds 500milliseconds You can also shorten the string by abbreviating the units to their first letter (milliseconds abbreviates to ms, microseconds to us): 4s 500ms The following are all equivalent: 1minute 3seconds A single number without any units specifies the value in milliseconds, but this functionality is deprecated and HPE recommends that you specify the relevant units. |
Default: | 1000ms |
Required: | No |
Configuration Section: | TaskName |
Example: | GPUBatchingDuration=1500ms
|
See Also: | GPUNumParallel |
|