GitHub user beltran opened a pull request:

    https://github.com/apache/tez/pull/23

    TEZ-3958: Add internal vertex priority information into the tez dag.d…

    …ot debug information
    
    This PR does the following:
    * Move `generateDAGVizFile` from `DagAppMaster` to `Utils` so it can be 
called from `DAGImpl`. The priorities are initialized after 
`DAGImpl.initializeDAG`, that why this call is done there.
    * There's still a call to `generateDAGVizFile` from `DagAppMaster` which 
would render the file but without the priorities. This file will be overwritten 
by the call in `DAGImpl`.
    * Creates methods `getPriorityLowLimit` and `getPriorityLowLimit` in 
`DAGScheduler`.
    * Creates `getDAGScheduler` in `DAG`.
    
    
    A sample `.dot` file looks like:
    ```
    digraph DAG_Iteration_0 {
    graph [ label="DAG_Iteration_0", fontsize=24, fontname=Helvetica];
    node [fontsize=12, fontname=Helvetica];
    edge [fontsize=9, fontcolor=blue, fontname=Arial];
    "DAG_Iteration_0.Sorter_Output" [ label = "Sorter[Output]", shape = "box" , 
color= "black"];
    "DAG_Iteration_0.Tokenizer" [ label = 
"Tokenizer[WordCount$TokenProcessor,\n priority=8,\n ]" , color= "black" ];
    "DAG_Iteration_0.Tokenizer" -> "DAG_Iteration_0.Summation" [ label = 
"[input=OrderedPartitionedKVOutput,\n output=OrderedGroupedKVInput,\n 
dataMovement=SCATTER_GATHER,\n schedulingType=SEQUENTIAL]" ];
    "DAG_Iteration_0.Tokenizer_Input" [ label = "Tokenizer[Input]", shape = 
"box" , color= "black"];
    "DAG_Iteration_0.Tokenizer_Input" -> "DAG_Iteration_0.Tokenizer" [ label = 
"Input [inputClass=MRInput,\n initializer=MRInputAMSplitGenerator]" ];
    "DAG_Iteration_0.Summation" [ label = 
"Summation[OrderedWordCount$SumProcessor,\n priority=11,\n ]" , color= "black" 
];
    "DAG_Iteration_0.Summation" -> "DAG_Iteration_0.Sorter" [ label = 
"[input=OrderedPartitionedKVOutput,\n output=OrderedGroupedKVInput,\n 
dataMovement=SCATTER_GATHER,\n schedulingType=SEQUENTIAL]" ];
    "DAG_Iteration_0.Sorter" [ label = "Sorter[OrderedWordCount$NoOpSorter,\n 
priority=14,\n ]" , color= "black" ];
    "DAG_Iteration_0.Sorter" -> "DAG_Iteration_0.Sorter_Output" [ label = 
"Output [outputClass=MROutput,\n committer=MROutputCommitter]" ];
    }
    ```

You can merge this pull request into a Git repository by running:

    $ git pull https://github.com/beltran/tez TEZ-3958

Alternatively you can review and apply these changes as the patch at:

    https://github.com/apache/tez/pull/23.patch

To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:

    This closes #23
    
----
commit daabde158fc58837a7bed059a79afaabc50e736c
Author: Jaume Marhuenda <jaumemarhuenda@...>
Date:   2018-06-27T20:55:38Z

    TEZ-3958: Add internal vertex priority information into the tez dag.dot 
debug information

----


---

Reply via email to