This is the documentation for Cloudera Manager 5.0.x. Documentation for other versions is available at Cloudera Documentation.

Viewing the Distribution of Task Attempts

The Task Distribution tab provides a graphical view of the performance of the Map and Reduce tasks that make up a job.

To display the task distribution metrics for a job:

  1. Do one of the following:
    • Select a MapReduce job from the Activities list.
    • Select a job from the Children tab of a Pig, Hive, or Oozie activity.
  2. Click the Task Distribution tab.

The chart that appears initially shows the distribution of Map Input Records by Duration; you can change the Y-axis to chart a number of different metrics.

You can use the Zoom to Duration button to zoom the Time Range Selector to the exact time range spanned by the activity whose tasks you are viewing.

The Task Distribution Chart

The Task Distribution chart shows the distribution of attempts according to their duration on the X-axis and a number of different metrics on the Y-axis. Each cell represents the number of tasks whose performance statistics fall within the parameters of the cell.

The Task Distribution chart is useful for detecting tasks that are outliers in your job, either because of skew, or because of faulty TaskTrackers. The chart can clearly show if some tasks deviate significantly from the majority of task attempts.

Normally, the distribution of tasks will be fairly concentrated. If, for example, some Reducers receive much more data than others, that will be represented by having two discrete sections of density on the graph. That suggests that there may be a problem with the user code, or that there's skew in the underlying data. Alternately, if the input sizes of various Map or Reduce tasks are the same, but the time it takes to process them varies widely, it might mean that certain TaskTrackers are performing more poorly than others.

You can click in a cell and see a list of the TaskTrackers that correspond to the tasks whose performance falls within the cell.

The X-axis show the task duration is seconds. From the drop-down you can chose different metrics for the Y-axis: Input or Output records or bytes for Map tasks, or the number of CPU seconds for the user who ran the job:

  • Map Input Records vs. Duration
  • Map Output Records vs. Duration
  • Map Input Bytes vs. Duration
  • Map Output Bytes vs. Duration
  • Map Total User CPU seconds vs. Duration
  • Reduce Input Records vs. Duration
  • Reduce Output Records vs. Duration
  • Reduce Total User CPU seconds vs. Duration

TaskTracker Hosts

To the right of the chart is a table that shows the TaskTracker hosts that processed the tasks in the selected cell, along with the number of task attempts each host executed.

You can select a cell in the table to view the TaskTracker hosts that correspond to the tasks in the cell.

  • The area above the TaskTracker table shows the type of task and range of data volume (or User CPUs) and duration times for the task attempts that fall within the cell.
  • The table itself shows the TaskTracker hosts that executed the tasks that are represented within the cell, and the number of task attempts run on that host.

Clicking a TaskTracker host name takes you to the Role Status page for that TaskTracker instance.

Page generated September 3, 2015.