Hi Devs, I have been working on adding support for cloud bursting in XBaya (see thread "0.7 release plan"). As the work is nearing completion, I am now thinking of a usecase to design a performance test to justify the value addition of Hadoop integration effort [1].
I need your assistance at this point for few things. 1. To come up with a good usecase that is based on some data-intensive scientific computing scenario. As I believe a scenario from any scientific paradigm would come in handy rather than going for a standard Hadoop scenario (like counting words) on a larger data set. 2. A data set to test the usecase 3. A run environment to set it up and do few rounds of testing. I am really curious to know what you think. Also, I hope Milinda would also have something to add. [1] - https://issues.apache.org/jira/browse/AIRAVATA-357 Thanks, Danushka
