Select Services

This section describes some of the services that can be run on a node.

Every installation requires services to manage jobs and applications. ResourceManager and NodeManager manage MapReduce version 2 and other applications that can run on YARN. In addition, MapR requires the ZooKeeper service to coordinate the cluster, and at least one node must run the CLDB service. The WebServer service is required if the browser-based MapR Control System will be used.

After you install MapR core, you can install ecosystem components that belong to a MapR Expansion Pack (MEP). A MEP provides a set of ecosystem components that work together. When a newer version or a revision to a component becomes available, the MEP version is updated to reflect the fact that an update was made. For details on the ecosystem components available in each MEP and the list of MEPs supported by your MapR cluster version, see MapR Expansion Packs (MEPs).

The following table shows some of the services that can be run on a node.

Service Category Service


Management Warden

Warden runs on every node, coordinating the node's contribution to the cluster. Warden is also responsible for managing the service state and its resource allocations on that node.

YARN NodeManager Hadoop YARN NodeManager service. The NodeManager manages node resources and monitors the health of the node. It works with the ResourceManager to manage YARN containers that run on the node.
MapR Core FileServer

FileServer is the MapR service that manages disk storage for MapR-FS and MapR-DB on each node.

MapR Core CLDB

Maintains the container location database (CLDB) service. The CLDB service coordinates data storage services among MapR-FS FileServer nodes, and access across MapR NFS gateways, and MapR clients.

MapR Core NFS

Provides read-write MapR Direct Access NFS™ access to the cluster, with full support for concurrent read and write access.

Storage MapR HBase Client

Provides access to MapR-DB binary tables via HBase APIs. Required on all nodes that will access table data in MapR-FS, typically all edge nodes for accessing table data. HBase API can also be accessed through the HBase Thrift and Rest Gateways.

YARN ResourceManager Hadoop YARN ResourceManager service. The ResourceManager manages cluster resources, and tracks resource usage and node health.
Management ZooKeeper

Internal service. Enables high availability (HA) and fault tolerance for MapR clusters by providing coordination.

YARN HistoryServer Archives MapReduce application metrics and metadata.
Management Web Server

Contains static MapR Control System UI pages.

Management Apiserver Allows you to perform cluster administration programmatically, and supports the MapR Control System (see Setting Up the MapR Control System).
OJAI Distributed Query Service Drill
Provides the distributed query service powered by Apache Drill for MapR-DB JSON. Supports the following functionality:
  • Advanced secondary index selection
  • Sorts on large data sets
  • Parallel query execution
See OJAI Distributed Query Service for more details about the service.
Application Hue Hue is the Hadoop User Interface that interacts with Apache Hadoop and its ecosystem components, such as Hive, Pig, and Oozie. It also provides interactive notebook access to Spark through Livy.
Application Pig

Pig is a high-level data-flow language and execution framework.

Application Hive

Hive is a data warehouse engine that supports SQL-like ad hoc querying and data summarization.

Application Flume

Flume is a service for piping and aggregating large amounts of log data

Application Oozie

Oozie is a workflow scheduler system for managing Hadoop jobs.

Application HCatalog

HCatalog provides applications with a table view of the MapR-FS layer of the cluster, expanding your options from read/write data streams to add-[Hive]-table operations such as get row and store row.

Application Cascading

Cascading is an application framework for analyzing and managing big data.

Application Myriad Myriad is an application framework that enables YARN applications and Mesos frameworks to run side-by-side while dynamically sharing cluster resources.

When using Myriad, the ResourceManager is deployed using Marathon, and NodeManager is run as a Mesos task.

Application Spark Spark is a processing engine for large datasets. While it can be deployed locally or standalone, the recommended deployment is on YARN. The application timeline server component provides a historical view of query details.
Application Sqoop

Sqoop is a library for transferring bulk data between Hadoop and relational databases.