Hadoop on Parallella

[Kanishk]

Hi,
I have a few queries as well as I would like to call all from the community who are interested/have crazy ideas in running Hadoop off a Parallella Board(s) in order for sharing information in this thread which could be helpful in the longterm and/or for other developers wanting to try their own M/R programs on this amazing platform.

But first, I would like to know about is the current status of any kind of support in Parallella's SDK for running Hadoop right out-of-box on a Parallella Board.

Also, as in the Ideas List (Sub-section: 1.4. Hadoop Many-Core Accelerator Support) this thing might be already under dev/test.

Thanks,
Kanishk Verma

[9600]

[Kanishk]

Hey

thanks for pointing it out and yes I am planning to try out a few MR programs.

Now since is out in the wild too (its been drawing quite an attention) I will be considering between Spark and Hadoop.

Regards,
Kanishk

[Kanishk]

well, spark is more or like HANA (in-memory proc.) and hence a deal breaker
Apache's Hadoop is in