pr feedback

Signed-off-by: Hemil Desai <hemild@nvidia.com>

Author: hemildesai

docs/source/guides/execution.md

@@ -226,13 +226,13 @@ As demonstrated in the examples, defining executors in Python offers great flexi
 
 #### RunAIExecutor
 
-The `RunAIExecutor` integrates with the Run:ai API to launch distributed jobs. It uses REST API calls to authenticate, identify the target project and cluster, and submit the job specification.
+The `RunAIExecutor` integrates with the NVIDIA Run:ai API to launch distributed jobs. It uses REST API calls to authenticate, identify the target project and cluster, and submit the job specification.
 
 Here's an example configuration:
 
 ```python
 def your_runai_executor(nodes: int, gpus_per_node: int, container_image: str):
-    # Ensure these are set correctly for your RunAI environment
+    # Ensure these are set correctly for your NVIDIA Run:ai environment
     # You might fetch these from environment variables or a config file
     base_url = "YOUR_RUNAI_API_ENDPOINT" # e.g., https://<cluster-name>.<domain>/api/v1
     app_id = "YOUR_RUNAI_APP_ID"
@@ -263,7 +263,7 @@ def your_runai_executor(nodes: int, gpus_per_node: int, container_image: str):
 
 ```
 
-For a complete end-to-end example using RunAI with NeMo, refer to the [NVIDIA RunAI NeMo End-to-End Workflow Example](https://docs.nvidia.com/dgx-cloud/run-ai/latest/nemo-e2e-example.html).
+For a complete end-to-end example using NVIDIA Run:ai with NeMo, refer to the [NVIDIA Run:ai NeMo End-to-End Workflow Example](https://docs.nvidia.com/dgx-cloud/run-ai/latest/nemo-e2e-example.html).
 
 #### LeptonExecutor
 

docs/source/guides/ray.md

@@ -63,7 +63,7 @@ executor = KubeRayExecutor(
     ],
     # Optional tweaks ----------------------------------------------------
     reuse_volumes_in_worker_groups=True,          # mount PVCs on workers too
-    spec_kwargs={"schedulerName": "runai-scheduler"},  # e.g. Run:ai
+    spec_kwargs={"schedulerName": "runai-scheduler"},  # e.g. NVIDIA Run:ai
     volume_mounts=[{"name": "workspace", "mountPath": "/workspace"}],
     volumes=[{
         "name": "workspace",

nemo_run/core/execution/runai.py

@@ -41,12 +41,12 @@ class RunAIState(Enum):
 @dataclass(kw_only=True)
 class RunAIExecutor(Executor):
     """
-    Dataclass to configure a RunAI Executor.
+    Dataclass to configure a NVIDIA Run:ai Executor.
 
-    This executor integrates with a RunAI cloud endpoint for launching jobs
+    This executor integrates with a NVIDIA Run:ai cloud endpoint for launching jobs
     via a REST API. It acquires an auth token, identifies the project/cluster,
     and launches jobs with a specified command. It can be adapted to meet user
-    authentication and job-submission requirements on RunAI.
+    authentication and job-submission requirements on NVIDIA Run:ai.
     """
 
     base_url: str
@@ -170,7 +170,7 @@ def move_data(self, token: str, project_id: str, cluster_id: str, sleep: float =
         workload_id = resp_json["workloadId"]
         status = RunAIState(resp_json["actualPhase"])
 
-        logger.info(f"Successfully created data movement workload {workload_id} on RunAI")
+        logger.info(f"Successfully created data movement workload {workload_id} on NVIDIA Run:ai")
 
         while status in [
             RunAIState.PENDING,
@@ -190,7 +190,7 @@ def move_data(self, token: str, project_id: str, cluster_id: str, sleep: float =
         resp = self.delete_workload(token, workload_id)
         if resp.status_code >= 200 and resp.status_code < 300:
             logger.info(
-                "Successfully deleted data movement workload %s on RunAI with response code %d",
+                "Successfully deleted data movement workload %s on NVIDIA Run:ai with response code %d",
                 workload_id,
                 resp.status_code,
             )
@@ -315,7 +315,7 @@ def cancel(self, job_id: str):
         response = requests.get(url, headers=headers)
         if response.status_code >= 200 and response.status_code < 300:
             logger.info(
-                "Successfully cancelled job %s on RunAI with response code %d",
+                "Successfully cancelled job %s on NVIDIA Run:ai with response code %d",
                 job_id,
                 response.status_code,
             )
@@ -330,7 +330,7 @@ def cancel(self, job_id: str):
     @classmethod
     def logs(cls: Type["RunAIExecutor"], app_id: str, fallback_path: Optional[str]):
         logger.warning(
-            "Logs not available for RunAIExecutor based jobs. Please visit the cluster UI to view the logs."
+            "Logs not available for NVIDIA Run:ai Executor based jobs. Please visit the cluster UI to view the logs."
         )
 
     def cleanup(self, handle: str): ...

nemo_run/run/torchx_backend/schedulers/runai.py

@@ -30,10 +30,10 @@
 from nemo_run.core.serialization.zlib_json import ZlibJSONSerializer
 from nemo_run.run.torchx_backend.schedulers.api import SchedulerMixin
 
-# Local placeholder for storing RunAI job states
+# Local placeholder for storing NVIDIA Run:ai job states
 RUNAI_JOB_DIRS = os.path.join(get_nemorun_home(), ".runai_jobs.json")
 
-# Example mapping from some RunAI statuses to the TorchX AppState
+# Example mapping from some NVIDIA Run:ai statuses to the TorchX AppState
 RUNAI_STATES: dict[RunAIState, AppState] = {
     RunAIState.CREATING: AppState.PENDING,
     RunAIState.INITIALIZING: AppState.SUBMITTED,
@@ -57,7 +57,7 @@
 @dataclass
 class RunAIRequest:
     """
-    Wrapper around the torchx AppDef and the RunAI executor.
+    Wrapper around the torchx AppDef and the NVIDIA Run:ai executor.
     This object is used to store job submission info for the scheduler.
     """
 
@@ -87,7 +87,7 @@ def _submit_dryrun(  # type: ignore
         cfg: Executor,
     ) -> AppDryRunInfo[RunAIRequest]:
         assert isinstance(cfg, RunAIExecutor), (
-            f"{cfg.__class__} not supported for RunAICloud scheduler."
+            f"{cfg.__class__} not supported for NVIDIA Run:ai Cloud scheduler."
         )
         executor = cfg
 
@@ -101,12 +101,12 @@ def _submit_dryrun(  # type: ignore
         return AppDryRunInfo(
             RunAIRequest(app=app, executor=executor, cmd=cmd, name=role.name),
             # Minimal function to show the config, if any
-            lambda req: f"RunAI job for app: {req.app.name}, cmd: {' '.join(cmd)}, executor: {executor}",
+            lambda req: f"NVIDIA Run:ai job for app: {req.app.name}, cmd: {' '.join(cmd)}, executor: {executor}",
         )
 
     def schedule(self, dryrun_info: AppDryRunInfo[RunAIRequest]) -> str:
         """
-        Launches a job on RunAI using the RunAIExecutor. Returns an app_id
+        Launches a job on NVIDIA Run:ai using the RunAIExecutor. Returns an app_id
         used by TorchX for subsequent queries/cancellations.
         """
         req = dryrun_info.request
@@ -119,7 +119,7 @@ def schedule(self, dryrun_info: AppDryRunInfo[RunAIRequest]) -> str:
         # We'll call it without additional parameters here.
         job_id, status = executor.launch(name=req.name, cmd=req.cmd)
         if not job_id:
-            raise RuntimeError("Failed scheduling run on RunAI: no job_id returned")
+            raise RuntimeError("Failed scheduling run on NVIDIA Run:ai: no job_id returned")
 
         # Example app_id format:
         # <experiment_id>___<role-name>___<job_id>
@@ -176,7 +176,7 @@ def describe(self, app_id: str) -> Optional[DescribeAppResponse]:
 
     def _cancel_existing(self, app_id: str) -> None:
         """
-        Cancels the job by calling the RunAIExecutor's cancel method.
+        Cancels the job by calling the NVIDIA Run:ai Executor's cancel method.
         """
         stored_data = _get_job_dirs()
         job_info = stored_data.get(app_id)