main/scripts/getGpusResources.sh

0001 #!/usr/bin/env bash
0002
0003 #
0004 # Licensed to the Apache Software Foundation (ASF) under one or more
0005 # contributor license agreements.  See the NOTICE file distributed with
0006 # this work for additional information regarding copyright ownership.
0007 # The ASF licenses this file to You under the Apache License, Version 2.0
0008 # (the "License"); you may not use this file except in compliance with
0009 # the License.  You may obtain a copy of the License at
0010 #
0011 #    http://www.apache.org/licenses/LICENSE-2.0
0012 #
0013 # Unless required by applicable law or agreed to in writing, software
0014 # distributed under the License is distributed on an "AS IS" BASIS,
0015 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
0016 # See the License for the specific language governing permissions and
0017 # limitations under the License.
0018 #
0019
0020 # This script is a basic example script to get resource information about NVIDIA GPUs.
0021 # It assumes the drivers are properly installed and the nvidia-smi command is available.
0022 # It is not guaranteed to work on all setups so please test and customize as needed
0023 # for your environment. It can be passed into SPARK via the config
0024 # spark.{driver/executor}.resource.gpu.discoveryScript to allow the driver or executor to discover
0025 # the GPUs it was allocated. It assumes you are running within an isolated container where the
0026 # GPUs are allocated exclusively to that driver or executor.
0027 # It outputs a JSON formatted string that is expected by the
0028 # spark.{driver/executor}.resource.gpu.discoveryScript config.
0029 #
0030 # Example output: {"name": "gpu", "addresses":["0","1","2","3","4","5","6","7"]}
0031
0032 ADDRS=`nvidia-smi --query-gpu=index --format=csv,noheader | sed -e ':a' -e 'N' -e'$!ba' -e 's/\n/","/g'`
0033 echo {\"name\": \"gpu\", \"addresses\":[\"$ADDRS\"]}