change: set btl_vader_single_copy_mechanism to none to avoid Read -1 …

…Warning messages (#95)
aws · Feb 2, 2021 · 0523a42 · 0523a42
1 parent a14e8f8
commit 0523a42
Show file tree

Hide file tree

Showing 2 changed files with 9 additions and 0 deletions.
diff --git a/src/sagemaker_training/smdataparallel.py b/src/sagemaker_training/smdataparallel.py
@@ -129,6 +129,9 @@ def _get_mpirun_command(
             "orte_abort_on_non_zero_status",
             "1",
             "-mca",
+            "btl_vader_single_copy_mechanism",
+            "none",
+            "-mca",
             "plm_rsh_num_concurrent",
             str(num_hosts),
             "-x",

diff --git a/test/unit/test_smdataparallel.py b/test/unit/test_smdataparallel.py
@@ -90,6 +90,9 @@ def test_smdataparallel_run_multi_node_python(
                 "orte_abort_on_non_zero_status",
                 "1",
                 "-mca",
+                "btl_vader_single_copy_mechanism",
+                "none",
+                "-mca",
                 "plm_rsh_num_concurrent",
                 str(num_hosts),
                 "-x",
@@ -189,6 +192,9 @@ def test_smdataparallel_run_single_node_python(
                 "orte_abort_on_non_zero_status",
                 "1",
                 "-mca",
+                "btl_vader_single_copy_mechanism",
+                "none",
+                "-mca",
                 "plm_rsh_num_concurrent",
                 str(num_hosts),
                 "-x",