SOAfaces Powers up Hadoop
Running and tracking Hadoop jobs takes a little extra effort in order to seamlessly integrate with a remote Hadoop environment. With release 2.6.0, it is now easier to compose, launch, and track Hadoop jobs remotely. This brings the full power of JobServer and soafaces to hadoop users. Hadoop users now can schedule their jobs and track them all the way through execution using a set of web based interfaces that provide role/user management, security and audit trail reporting among many other features.
One very nice feature with using soafaces and JobServer, is that you can chain multiple Hadoop jobs together and send the results of one Hadoop job into the input of another Hadoop job. You can transfer files and content in out and out of HDFS and orchestrate and chain of Hadoop jobs to complete a Hadoop workflow.
Release 2.6.0 includes some key new features like application and runtime scope cookies that let a Tasklet manage state between job runs enabling better coordination with third party systems like Hadoop.
This release also includes two built in soafaces components that let you report on your Hadoop environment and launch and track any Hadoop job. These components are bundled with the other BeanSoup components that come as part of the sample components with soafaces.
With these components, you can build and customize your own GWT GUI interfaces for each Hadoop job you have. soafaces provides a standard component/Tasklet you can you use out of the box to run your custom Hadoop jobs. And you can tailor the code with your own custom logic to provide specialized GWT reporting for each unique Hadoop job, for example.
Use these links to try soafaces, download JobServer and start managing your Big Data from the comfort of your JobServer Workbench.
(Note: Opinions expressed in this article and its replies are the opinions of their respective authors and not those of DZone, Inc.)