This is documentation for MapR Version 5.0. You can also refer to MapR documentation for the latest release.

Skip to end of metadata
Go to start of metadata

You can edit the hadoop_version file on a MapR client node to configure the default MapReduce mode for all jobs and application that you submit from the client node. The mode that you set on the client overrides the MapReduce mode of the cluster.

For information about how this impacts ecosystem clients, see Managing the MapReduce Mode for Ecosystem Components.

To set the MapReduce mode on the client node:

  1. Open the hadoop_version file in the following location: /opt/mapr/conf/
  2. Edit the default_mode parameter.
    The following values are valid for the default_mode parameter:

    SettingsDescription
    default_mode=classic

    Specifies that the client submits MapReduce v1 jobs to the cluster.

    default_mode=yarnSpecifies that the client submits MapReduce v2 applications to the cluster.
    Icon

    The MapReduce mode setting on the client node enables all of the hadoop command that you run on this client to operate using the hadoop version associated with that mode. For example, if you set the MapReduce mode to yarn, when you run hadoop classpath the classpath associated with hadoop 2.x displays.

  • No labels