Embrace Windows Hadoop--hdinsight

The big data is so real, it's getting closer, and you no longer need the complicated Linux operation to embrace Windows Hadoop--hdinsight. Hdinsight is an implementation of the 100% compatible http://www.aliyun.com/zixun/aggregation/14417.html ">apache Hadoop" on a Windows platform. and Microsoft provides it with complete technical support, and what are we going to do to get into the big data world together?

Currently available hdinsight are divided into two versions:

On-premises version, Hdinsight Server

Cloud version, Hdinsight Service

Since the Hdinsight service is not yet open and requires an invitation, we use Hdinsight server to experience the power of large data.

Installation article
=======================

At present, the Hdinsight installation is implemented through the Web Platform installer, you do not care about all the dependencies, all of these installers have been done for you. One thing we have to point out is that in fact the installer installs two major components, Hortonworks Data Platform for Windows and Microsoft Hdinsight, and the rest of the required components, such as IIS, are automatically installed by Setup. By default, installing Hdinsight on a clean Windows Server 2012 installs the following components:

Let's discuss separately what they are:

Hortonworks Data Platform Installer

This MIS contains the core implementations of Hadoop (such as Map/reduce, HDFS), and some other Apache projects in the Hadoop ecosystem, the following is a complete list:

· Map Reduce

· HDFS

· Hive

· Pig

· Hcatalog

Each project is packaged into a zip file, and they all contain a PowerShell script to complete the installation.

Microsoft Hdinsight Installer

This MSI contains some of the components of Microsoft's implementation that currently contain:

· Hdinsight Dashboard

· Sqoop

· Isotope.js

· Getting Started content

These components are packaged in a manner similar to the way the Hadoop project was organized above. Also, it provides a PowerShell script that can be used to initialize the installation, such as the service used to start the Hadoop component.