Remove CUDAService numberOfStreamsPerDevice as unsupported #282
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Currently the
SwitchProducer(CUDA)
(see #100) decides between CPU and GPU algorithm at the beginning of the job. This approach does not allow supporting thenumberOfStreamsPerDevice
parameter ofCUDAService
. Therefore this PR suggests to remove that functionality fromCUDAService
.If the feature is wanted back, the SwitchProducer would have to be enhanced to do the device type decisions event-by-event (see #281).