Let the user configure how many cuda/hip streams or sycl queues they would like to have in the resource implementation.
Let the user configure how many cuda/hip streams or sycl queues they would like to have in the resource implementation.