Profiling Tool - Jar Usage

The Profiling tool can be run in as a java cmd in three different ways if you aren’t using the CLI tool:

There are 3 modes of operation for the Profiling tool:

For sample execution commands, refer to the examples section.

Prerequisites

• Java 8+

• Spark event log(s) from Spark 2.0 or above version. Supports both rolled and compressed event logs with .lz4, .lzf, .snappy and .zstd suffixes as well as Databricks-specific rolled and compressed(.gz) event logs.

• The tool requires the Spark 3.x+ jars to be able to run but it doesn’t need an Apache Spark runtime. If you don’t already have Spark 3.x+ installed, you can download the Apache Spark Distribution to any machine and include the jars in the classpath.

• This tool parses the Spark CPU event log(s) and creates an output report. Acceptable inputs are either individual or multiple event logs files or directories containing spark event logs in the local filesystem, HDFS, S3, ABFS, GCS or mixed. If you want to point to the local filesystem be sure to include prefix file: in the path. If any input is a remote file path or directory path, then you need to the connector dependencies to be on the classpath

  • On-prem HDFS
  • GCS
  • S3 :sync: s3-key
  • ABFS

  Include $HADOOP_CONF_DIR in classpath

  Sample showing Java’s classpath
  

  Copy

  Copied!

              
  
              
  -cp ~/rapids-4-spark-tools_2.12-<version>.jar:$SPARK_HOME/jars/*:$HADOOP_CONF_DIR/
          

  
  

  Download the gcs-connector-hadoop3-<version>-shaded.jar and follow the instructions to configure Hadoop/Spark.

  Download the matched jars based on the Hadoop version

  • hadoop-aws-<version>.jar

  • aws-java-sdk-<version>.jar

  In $SPARK_HOME/conf, create hdfs-site.xml with below AWS S3 keys inside:

  Copy

  Copied!

              
  
              
  <?xml version="1.0"?>
  <configuration>
     <property>
        <name>fs.s3a.access.key</name>
        <value>xxx</value>
     </property>
     <property>
        <name>fs.s3a.secret.key</name>
        <value>xxx</value>
     </property>
  </configuration>
          

  You can test your configuration by including the above jars in the -jars option to spark-shell or spark-submit

  Refer to the Hadoop-AWS doc on more options about integrating Hadoop-AWS module with S3.

  • Download the matched jar based on the Hadoop version hadoop-azure-<version>.jar.

  • The simplest authentication mechanism is to use account-name and account-key.Refer to the Hadoop-ABFS support doc on more options about integrating Hadoop-ABFS module with ABFS.

  
  

Getting the Tools Jar

Profiling Tool Options

            

            
Profiling tool for the RAPIDS Accelerator and Apache Spark

Usage: java -cp rapids-4-spark-tools_2.12-<version>.jar:$SPARK_HOME/jars/*
       com.nvidia.spark.rapids.tool.profiling.ProfileMain [options]
       <eventlogs | eventlog directories ...>

  -a, --auto-tuner                Toggle AutoTuner module.
      --combined                  Collect mode but combine all applications into
                                  the same tables.
  -c, --compare                   Compare Applications (Note this may require
                                  more memory if comparing a large number of
                                  applications). Default is false.
      --csv                       Output each table to a CSV file as well
                                  creating the summary text file.
  -d, --driverlog <arg>           Specifies the name of a driver log file that
                                  the profiling tool is to process. The tool
                                  identifies any invalid operations in the log
                                  and writes them to a .csv file. When
                                  --driverlog is specified, the eventlog
                                  parameter is optional.
  -f, --filter-criteria  <arg>    Filter newest or oldest N eventlogs based on
                                  application start timestamp for processing.
                                  Filesystem based filtering happens before
                                  application based filtering (see start-app-time).
                                  for example, 100-newest-filesystem (for processing newest
                                  100 event logs). For example, 100-oldest-filesystem (for
                                  processing oldest 100 event logs).
  -g, --generate-dot              Generate query visualizations in DOT format.
                                  Default is false
      --generate-timeline         Write an SVG graph out for the full
                                  application timeline.
  -m, --match-event-logs  <arg>   Filter event logs whose filenames contain the
                                  input string
  -n, --num-output-rows  <arg>    Number of output rows for each Application.
                                  Default is 1000
      --num-threads  <arg>        Number of thread to use for parallel
                                  processing. The default is the number of cores
                                  on host divided by 4.
  -o, --output-directory  <arg>   Base output directory. Default is current
                                  directory for the default filesystem. The
                                  final output will go into a subdirectory
                                  called rapids_4_spark_profile. It will
                                  overwrite any existing files with the same
                                  name.
  -p, --print-plans               Print the SQL plans to a file named
                                  'planDescriptions.log'.
                                  Default is false.
  -s, --start-app-time  <arg>     Filter event logs whose application start
                                  occurred within the past specified time
                                  period. Valid time periods are
                                  min(minute),h(hours),d(days),w(weeks),m(months).
                                  If a period isn't specified it defaults to
                                  days.
  -t, --timeout  <arg>            Maximum time in seconds to wait for the event
                                  logs to be processed. Default is 24 hours
                                  (86400 seconds) and must be greater than 3
                                  seconds. If it times out, it will report what
                                  it was able to process up until the timeout.
  -w, --worker-info  <arg>        File path containing the system information of
                                  a worker node. It's assumed that all workers
                                  are homogenous. It requires the AutoTuner to
                                  be enabled. Default is ./worker_info.yaml
  -h, --help                      Show help message

 trailing arguments:
  eventlog (optional)   Event log filenames (space separated) or directories
                        containing event logs. For example, s3a://<BUCKET>/eventlog1
                        /path/to/eventlog2. At least one eventlog or a driver
                        log must be specified; thus an eventlog parameter is
                        required if the --driverlog option isn't specified.
        

Tuning Spark Properties For GPU Clusters

Currently, the Auto-Tuner calculates a set of configurations that impact the performance of Apache Spark apps executing on GPU. Those calculations can leverage cluster information (for example, memory, cores, Spark default configurations) as well as information processed in the application event logs. The tool also will recommend settings for the application assuming that the job will be able to use all the cluster resources (CPU and GPU) when it’s running. The values loaded from the app logs have higher precedence than the default configs.

To run the Auto-Tuner, enable the auto-tuner flag and optionally pass a valid --worker-info <FILE_PATH>. The Auto-Tuner needs to learn the system properties of the worker nodes that run application code in the cluster. The argument FILE_PATH can either be local or remote file (that is, HDFS).

If the --worker-info argument isn’t supplied, then the Auto-Tuner will only recommend tuned settings based on the job event log and not on any cluster or worker information since that isn’t available.

Template of the worker information file in “yaml” format

            

            
system:
  numCores: 32
  memory: 212992MiB
  numWorkers: 5
gpu:
  memory: 15109MiB
  count: 4
  name: T4
softwareProperties:
  spark.driver.maxResultSize: 7680m
  spark.driver.memory: 15360m
  spark.executor.cores: '8'
  spark.executor.instances: '2'
  spark.executor.memory: 47222m
  spark.executorEnv.OPENBLAS_NUM_THREADS: '1'
  spark.scheduler.mode: FAIR
  spark.sql.cbo.enabled: 'true'
  spark.ui.port: '0'
  spark.yarn.am.memory: 640m
        

Processing Spark Event Logs

1. The tool reads the log files and process them in-memory. So the heap memory should be increased when processing large volume of events. It’s recommended to pass VM options -Xmx10g and adjust according to the number-of-apps / size-of-logs being processed.

   Copy

   Copied!

               
   
               
   export JVM_HEAP=-Xmx10g
           

2. Examples running the tool on the following environments

   • Local Filesystem
   • On-prem HDFS

   • Extract the Spark distribution into a local directory if necessary.

   • Either set SPARK_HOME to point to that directory or just put the path inside of the classpath java -cp toolsJar:$SPARK_HOME/jars/*:... when you run the Qualification tool.

   Copy

   Copied!

               
   
               
   java ${JVM_HEAP} \
        -cp rapids-4-spark-tools_2.12-<version>.jar:$SPARK_HOME/jars/* \
        com.nvidia.spark.rapids.tool.profiling.ProfileMain [options] \
        <eventlogs | eventlog directories ...>
           

   Copy

   Copied!

               
   
               
   java ${JVM_HEAP} \
        -cp rapids-4-spark-tools_2.12-<version>.jar:$SPARK_HOME/jars/* \
        com.nvidia.spark.rapids.tool.profiling.ProfileMain \
        /usr/logs/app-name1
           

   Example running on files in HDFS: (include $HADOOP_CONF_DIR in classpath). Note, on an HDFS cluster, the default filesystem is likely HDFS for both the input and output so if you want to point to the local filesystem be sure to include file: in the path.

   Copy

   Copied!

               
   
               
   java ${JVM_HEAP} \
        -cp rapids-4-spark-tools_2.12-<version>.jar:$SPARK_HOME/jars/*:$HADOOP_CONF_DIR/ \
        com.nvidia.spark.rapids.tool.profiling.ProfileMain  /eventlogDir
           

Processing Driver Logs

The Profiling tool can process GPU a driver log as well as CPU and GPU event logs. When the Profiling tool processes a driver log, it generates a .csv file that lists unsupported operators.

You inform the Profiling tool of a GPU driver log with the command line option --driverlog. The option has one required argument, specifying the pathname of a driver log file. You may specify just one driver log file per a single run.

A single run of the Profiling tool may process CPU/GPU event logs, a GPU driver log, or both.

Please refer to Processing event logs section for instructions on accessing the driver log existing on remote and local filesystems.

            

            
java ${JVM_HEAP} \
      -cp rapids-4-spark-tools_2.12-<version>.jar:$SPARK_HOME/jars/* \
      com.nvidia.spark.rapids.tool.profiling.ProfileMain  \
      --driverlog /path_to_driverlog \
      /eventlog
        

Collection Modes

Example running Profiling tool with different collections modes:

            

            
java -cp rapids-4-spark-tools_2.12-<version>.jar:$SPARK_HOME/jars/* \
     com.nvidia.spark.rapids.tool.profiling.ProfileMain [options] \
     <eventlogs | eventlog directories ...>
        

            

            
java -cp rapids-4-spark-tools_2.12-<version>.jar:$SPARK_HOME/jars/* \
     com.nvidia.spark.rapids.tool.profiling.ProfileMain --combined \
     <eventlogs | eventlog directories ...>
        

            

            
java -cp rapids-4-spark-tools_2.12-<version>.jar:$SPARK_HOME/jars/* \
     com.nvidia.spark.rapids.tool.profiling.ProfileMain --compare \
     <eventlogs | eventlog directories ...>