Hadoop on Azure. Big Data made easy – Part 1

Microsoft’s end-to-end roadmap for Big Data embraces Apache Hadoop™ by distributing enterprise class Hadoop-based flavors that will be available to both Windows Server and Windows Azure and as you probably know Hadoop on Azure is currently in Developer Preview (hadooponazure.com). The most impressive feature that you can find in the Hadoop on Azure  service is the extremely easy user interaction that has been implemented, for example take a look at how to create a new Hadoop on Azure Cluster:

1. You need to specify a DNS name to access your cluster. Select the Cluster size and introduce a user name and password.
2. When it’s all OK you just need to click on “Request Cluster”
3.- The allocation process will start and all that you need to do is wait for few minutes.
4.- When you have successfully completed the allocation process, you get the cluster control panel where you can create jobs, use the interactive Javascript or Hive console, start a Remote Desktop to the Head Node, open ports for FTPS or ODBC, manage your data (including configuring Azure or S3 storage), check your Billing History, Job History, use the samples or download the Hive add-in for Excel.

And that’s it. I’m sure things may change (this is the Developer Preview, so any change could happen), but I’m sure that will be to improve the experience.

Comments

• Anonymous
  February 13, 2012
  Good one - looking forward to its release.

• Anonymous
  March 20, 2012
  handy article... thank you for sharing... one typo... "hadooponazure.com" points wrong link