is it possible to stream data to global memory while a kernel is executing on the device ?
No. See the FAQ “Programming Questions” #4 [url=“http://forums.nvidia.com/index.php?showtopic=36286”]http://forums.nvidia.com/index.php?showtopic=36286[/url] or the many other forum posts on this topic for details.