Are worker and worker_node used interchangeably?

Ray Clusters
1

How severe does this issue affect your experience of using Ray?

  • Medium: It contributes to significant difficulty to complete my task, but I can work around it.

Very confused on the workers vs worker_nodes in Ray. Would appreciate some help!!

  • When lauching a ray cluster using setup_ray_cluster(num_cpus_worker_node=2, max_worker_nodes=16) .I ran it on one databricks instance with 32 cores and 1 gpu, it actually did create 16 worker_nodes. So this seems to suggest one worker_node is actually a worker process? Or it does create a node and we can further define worker process on each worker_node?
  • When using ScalingConfig(num_workers=1), here the num_workers is number of worker processes or number of worker nodes? what is the relationship between a worker and a worker_node?
  • When using ScalingConfig(num_workers=4, use_gpu=True) , the warning tells me I am requesting one cpu and four gpu. I am very confused by why
2

what ray version on databricks are you running?