Win7+ubuntu dual system installation and Hadoop pseudo-distributed installation

First install dual system for pseudo-distributed experiment, install Win7+ubuntu dual system:

1. Right click on "My Computer" to go to "manage", double click "Storage", then double click "Disk Management", in the D disk position right click "Compress volume", separate a size of 50G disk space, then format, then delete the volume, as used to install Ubuntu system.

2, download install EASYBCD software, create a new Neo grub boot, and then click on the settings, add a statement:

Title Install Ubuntu 12.04 LTS

Root (hd0,5)

Kernel (hd0,5)/vmlinuz boot=casper iso-scan/filename=/ubuntu-12.04-desktop-i386.iso ro quiet splash locale=zh_cn. UTF-8

INITRD (hd0,5)/initrd.lz

Note: 5 is on behalf of D, because my image is on the D drive, 1-4 for the primary partition and the C drive.

3, boot into the graphical interface to install Ubuntu can be.

Next into the Ubuntu system, first the current user into the root user, open the terminal to enter the following command, sudo gedit/etc/passwd find the current user name line, change all 1000 to 0, save exit, and then log off the user, log in again. If you don't want to do this, you can also add sudo before using the command.

4. Install Java

Download and unzip the jdk-7u51-linux-i586.tar.gz to/usr directory, rename the folder to the JVM, open the terminal, enter the command vim/etc/profile edit the environment variable, and add the following statement at the end:

e xport JAVA_HOME=/USR/JVM

Export classpath=.: $JAVA _home/lib/dt.jar: $JAVA _home/lib/tools.jar: $JAVA _home/lib: $CLASSPATH

Export path= $JAVA _home/bin: $PATH

Exit after saving, and then enter Source/etc/profile to make the settings effective.

5, install SSH and do password-free connection

Open terminal, enter command: apt-get install SSH. After the installation is complete, set up SSH login without password:

6. Then install Hadoop

Download and unzip the hadoop1.2.1.tar.gz to/usr, rename it to Hadoop, go to the hadoop/conf directory, first edit hadoop-env.sh, and add the Java installation path inside.

Then edit core-site.xml, and note that the TMP directory needs to be created in advance. Second, Fs.default.name is the location of the specified NN:

Then the editor hdfs-site.xml. This is the configuration replication factor, because it is now a pseudo-distribution, so there is only one DN, so it is 1.

The second is mapred-site.xml. The Mapred.job.tracker is the location of the specified JT.

Save exit. Then the Namenode is formatted, open the terminal, navigate to the Hadoop directory, enter the command: Hadoop Namenode-format Enter, see that the format is successful. If you add the bin directory of Hadoop to an environment variable, you can run the Hadoop command directly.

Then we came to the most exciting moment! Run hadoop!! After you run the JPS command, you can see that there should be 5 processes.

You can access the 50070 port of localhost, which is the web operator interface of NN.

You can also access the 50030 port of localhost. This is JT's web monitoring interface.

Here, congratulations! Hadoop pseudo-distribution is complete!!

Win7+ubuntu dual system installation and Hadoop pseudo-distributed installation