Construct the platform for Big Data project
Build HardDisk RAID0
This paragraph describes the HDFS configure progress on Azure Cloud Server. You need have a azure account to login in and purchase the platform you will perform on. Here is my platform construction details to share with you.
- Unbuntu 14.0 LTS
lvm2
- command –lvcreate
- Configure the raid 0
- Login in the url: http://manage.windowsazure.com
- Attached/Added 4 empty disks
- Create logical volume ( striped disks)
1 | $ sudo apt-get update |
- Automatic Mount configure (fstab)
1 | sudo blkid |
1 | $ time sh -c "sudo dd if=/dev/zero of=/datasource/temp bs=4k count=2000000 && sync" |
HDFS
- Login in http://manage.windowsazure.com, build the server on Azure
- Set the host: example-hadoop.cloudapp.net
- Create VM (Ubuntu 14 LTS), with 8 cores and 14GB RAM
- example-data01, example-data02, example-data03, example-data04
- Hadoop FAQ
- ssh-keygen
1 | admin@example-data01 $ cd .ssh |
Download the Hadoop Binary on each node1
2
3
4$ Install java on ubuntu
$ wget http://mirrors.cnnic.cn/apache/hadoop/common/stable/hadoop-2.7.1.tar.gz
$ tar xvf hadoop-2.7.1.tar.gz
$ mv hadoop-2.7.1 hadoop