Managing the Impala Llama ApplicationMaster

The Impala Llama ApplicationMaster (Llama) reserves and releases YARN-managed resources for Impala, thus reducing resource management overhead when performing Impala queries. Llama is used when you want to enable integrated resource management.

By default, YARN allocates resources bit-by-bit as needed by MapReduce jobs. Impala needs all resources available at the same time, so that intermediate results can be exchanged between cluster nodes, and queries do not stall partway through waiting for new resources to be allocated. Llama is the intermediary process that ensures all requested resources are available before each Impala query actually begins.

For more information about Llama, see Llama - Low Latency Application MAster.

For information on enabling Llama high availability, see Llama High Availability.

Enabling Integrated Resource Management Using Cloudera Manager

Minimum Required Role: Cluster Administrator (also provided by Full Administrator)

The Enable Integrated Resource Management wizard enables cgroups for the all the hosts in the cluster running Impala and YARN, adds one or more Llama roles to the Impala service, and configures the Impala and YARN services.

  1. Start the wizard using one of the following paths:
    • Cluster-level
      1. Select Clusters > Cluster name > Dynamic Resource Pools.
      2. In the Status section, click Enable.
    • Service-level
      1. Go to the Impala service.
      2. Select Actions > Enable Integrated Resource Management.
    The Enable Integrated Resource Management wizard starts and displays information about resource management options and the actions performed by the wizard.
  2. Click Continue.
  3. Leave the Enable Cgroup-based Resource Management checkbox checked and click Continue.
  4. Click the Impala Llama ApplicationMaster Hosts field to display a dialog for choosing Llama hosts.
    The following shortcuts for specifying hostname patterns are supported:
    • Range of hostnames (without the domain portion)
      Range Definition Matching Hosts
      10.1.1.[1-4] 10.1.1.1, 10.1.1.2, 10.1.1.3, 10.1.1.4
      host[1-3].company.com host1.company.com, host2.company.com, host3.company.com
      host[07-10].company.com host07.company.com, host08.company.com, host09.company.com, host10.company.com
    • IP addresses
    • Rack name
  5. Specify or select one or more hosts and click OK.
  6. Click Continue. A progress screen displays with a summary of the wizard actions.
  7. Click Continue.
  8. Click Restart Now to restart the cluster and apply the configuration changes or click leave this wizard to restart at a later time.
  9. Click Finish.

Disabling Integrated Resource Management Using Cloudera Manager

Minimum Required Role: Cluster Administrator (also provided by Full Administrator)

The Enable Integrated Resource Management wizard enables cgroups for the all the hosts in the cluster running Impala and YARN, adds one or more Llama roles to the Impala service, and configures the Impala and YARN services.

  1. Start the wizard using one of the following paths:
    • Cluster-level
      1. Select Clusters > Cluster name > Dynamic Resource Pools.
      2. In the Status section, click Disable.
    • Service-level
      1. Go to the Impala service.
      2. Select Actions > Disable Integrated Resource Management.
    The Disable Integrated Resource Management wizard starts and displays information about resource management options and the actions performed by the wizard.

Configuring Llama Using Cloudera Manager

Minimum Required Role: Configurator (also provided by Cluster Administrator, Full Administrator)

  1. Go to the Impala service.
  2. Click the Configuration tab.
  3. Select Scope > Impala Llama ApplicationMaster.
  4. Edit configuration properties.
  5. Click Save Changes to commit the changes.
  6. Restart the Llama role.