Need to analyze Tez DAG information particularly the critical path on HDInsight cluster
1) Connect to the HDInsight cluster with a Secure Shell (SSH) client (check Further Reading section below).
2) Run the following command at the command prompt:
hadoop jar /usr/hdp/current/tez-client/tez-job-analyzer-*.jar CriticalPath --saveResults --dagId <DagId> --eventFileName <DagData.zip>
3) List other analyzers that can be used for analyzing Tez DAG with the following command:
hadoop jar /usr/hdp/current/tez-client/tez-job-analyzer-*.jar
An example program must be given as the first argument.
Valid program names are:
ContainerReuseAnalyzer: Print container reuse details in a DAG
CriticalPath: Find the critical path of a DAG
LocalityAnalyzer: Print locality details in a DAG
ShuffleTimeAnalyzer: Analyze the shuffle time details in a DAG
SkewAnalyzer: Analyze the skew details in a DAG
SlowNodeAnalyzer: Print node details in a DAG
SlowTaskIdentifier: Print slow task details in a DAG
SlowestVertexAnalyzer: Print slowest vertex details in a DAG
SpillAnalyzer: Print spill details in a DAG
TaskConcurrencyAnalyzer: Print the task concurrency details in a DAG
VertexLevelCriticalPathAnalyzer: Find critical path at vertex level in a DAG