Hadoop学习笔记

[b][size=x-large]Hadoop介绍[/size][/b]

Hadoop是Google云计算框架的开源实现，是一个分布式存储和分布式计算的框架，主要包括HDFS和MapReduce的实现。

[b][size=large] HDFS[/size][/b]

HDFS由一个NameNode和多个DataNode组成，其中NameNode相当于系统的元数据存放地，它是Hadoop系统的神经中枢，而多个DataNode存储数据。

[b][size=large]MapReduce:分布式计算[/size][/b]

一个调用客户端由一个JobTracker代表，它将一个任务划分为多个子任务，每个子任务分别由一个TaskTracker负责。TaskTracker和DataNode在一起，本地数据本地计算。


[b][size=x-large]Hadoop的子项目[/size][/b]

[b][size=large]Avro™[/size][/b]

A data serialization system.


[b][size=large]Cassandra™:[/size][/b]

A scalable multi-master database with no single points of failure.

[b][size=large]Chukwa™: [/size][/b]

A data collection system for managing large distributed systems.

[b][size=large]HBase™:[/size][/b]

A scalable, distributed database that supports structured data storage for large tables.

[b][size=large]Hive™: [/size][/b]

A data warehouse infrastructure that provides data summarization and ad hoc querying.

[b][size=large]Mahout™: [/size][/b]

A Scalable machine learning and data mining library.

[b][size=large]Pig™: [/size][/b]

A high-level data-flow language and execution framework for parallel computation.

[b][size=large]ZooKeeper™:[/size][/b]

A high-performance coordination service for distributed applications.