Cascading for the Impatient, Part 1

[Paco Nathan]

We've got a new series of blog posts, called "Cascading for the
Impatient". The intent is to show brief/concise examples of Cascading
apps, starting from the smallest possible and progressing up to
something a bit more beefy, a TF-IDF implementation.

Here's "Part 1":
http://www.cascading.org/2012/07/02/cascading-for-the-impatient-part-1/

paco

[tom kern]

looking forward to the rest of the series as i have to implement a tf-idf implementation now too. will be interesting to see how cascading pro's are doing it ;)

Thomas

[Bertrand Dechoux]

Great. It would be nice to speak a bit about local mode for testing.
I found it one of the greatest feature of the version 2.

Bertrand

[Paco Nathan]

Thank you Bertrand!

I really appreciate the feedback about this code example. We're trying
to get it into shape to use for live demos.

In #7 of the "six-part" series, I was hoping to show local mode, plus
cmd line options to toggle some of the TDD features.

Paco

> --
> You received this message because you are subscribed to the Google Groups "cascading-user" group.
> To post to this group, send email to cascadi...@googlegroups.com.
> To unsubscribe from this group, send email to cascading-use...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/cascading-user?hl=en.
>

[John Watson]

Hello Paco,

I just did part 1, and the only difference I had was that when running on a single-node pseudo-distributed hadoop instance, I get the output split across part-00000 and part-00001, rather than all being in one partition file.  I don't know if that's expected, or I have something configured incorrectly somewhere (there seems to be a lot of opportunity for that with hadoop).

Great so far!

John Watson