Using a GPU is unrelated to Spark. You can run code that uses GPUs. This error indicates that something failed when you ran your code (GPU OOM?) and you need to investigate why.
On Wed, Nov 23, 2022 at 7:51 AM Vajiha Begum S A < vajihabegu...@maestrowiz.com> wrote: > Hi Sean Owen, > I'm using windows system with the NVIDIA Quadro K1200. > GPU memory 20GB, Intel Core: 8 core > installed - CUDAF 0.14 jar file, Rapid 4 Spark 2.12-22.10.0 jar file, CUDA > Toolkit 11.8.0 windows version. > Also installed- WSL 2.0 ( since I'm using windows system) > I'm running only single server, Master is localhost > > I'm trying to run pyspark code through Python idle. [But it showing > Winerror:10054, An existing connection was forcibly closed by the remote > host] > > I'm getting doubt that I can run pyspark code using GPU in the Windows > system? > Or is the only ubuntu/linux system the only option to run pyspark with GPU > set up. > Kindly give suggestions. Reply me back. >
