All Documents
Current Document

Content is empty

If you don't find the content you expect, please try another search term



Last updated:2020-07-17 11:54:10

Kingsoft Cloud MapReduce (KMR) is a one-stop open source big data platform built on Kingsoft Cloud. It provides a platform for the collection, storage, processing and presentation of massive data. KMR platform is fully compatible with the open source Hadoop/Spark community, improves the availability and stability of open source components, and implements the managed operation and maintenance, elastic management and security control of big data platform, enabling enterprises to focus on their business, improve the productivity, while reducing the cost and cycle of business operation.
KMR is a scalable general-purpose data computing and analysis platform that helps you quickly build a distributed data analysis system. KMR is built on two data computing frameworks: Apache Hadoop and Apache Spark, and it separates the computing and storage. KMR can also support the data transmission with other cloud data storage system and database system of Kingsoft Cloud (such as KS3).
KMR provides high scalability and elastic scalability, and eliminates Hadoop installation and deployment costs and management complexity, so you can focus on the data analysis and processing, not the infrastructure management. Any developer or company can perform large-scale data analysis and processing at low cost.
KMR Cluster consists of the master node and several core nodes.

  • Master node: The master node is used mainly for the cluster management, and to run the management process in Hadoop cluster, such as namenode, resource manager, jobhistory , etc. In addition, it tracks the execution status of each computing job, monitors the running status of the instance, and provides you with a portal for the cluster access. One KMR Cluster has only one master node.
  • Core node: A core node is used mainly to perform the cluster computing jobs. It is also the data storage node in the Distributed File System (HDFS), and corresponds to the slave node in Hadoop cluster. One KMR Cluster can have two or more core nodes.
    As per different business features, KMR provides a variety of node configurations (please refer to Pricing and selection of Kingsoft MapReduce), so you can choose right configuration according to your own needs, and increase or decrease the number of nodes dynamically as needed. In addition to Hadoop, KMR integrates many ecosystem components like Spark, Hive or Pig, so various computing jobs including MapReduce, Spark, Hive or Pig can be executed through the console or API.
On this page
Pure ModeNormal Mode

Pure Mode

Click to preview the document content in full screen