R help - Dec 2010 - Running scripts in hadoop

R-help group,

I'm looking for some assistance on using an R-script to read STDIN from
hadoop.
Example, say I have two tables. One is a student table, the other is a class
roster table (tables join on student_id). Student SAT score is in the
student table, whether the student passed or not is in the roster table.

So to determine if a student passed or failed based on their SAT score, I'd
run this:
SELECT student.SAT, roster.pass FROM student JOIN roster ON
student.student_id=roster.student_id

Easy enough, but now I want to pass the results of this query through an
R-script so I can perform analysis on this. How would I go about passing the
results of a query as a STDIN in hadoop (aka what would the syntax look
like)?

Thanks,

	[[alternative HTML version deleted]]

Also, in case one was wondering, the output would look something like:

1900 1
1850 1
1800 1
1700 0
1700 1
1680 1
1600 0
1500 0
2000 1
2100 1
2150 1
1750 1
1550 1
...



On Fri, Dec 24, 2010 at 5:28 PM, Joseph Regan <jregan@chitika.com> wrote:
> R-help group,
>
> I'm looking for some assistance on using an R-script to read STDIN from
> hadoop.
> Example, say I have two tables. One is a student table, the other is a
> class roster table (tables join on student_id). Student SAT score is in the
> student table, whether the student passed or not is in the roster table.
>
> So to determine if a student passed or failed based on their SAT score,
I'd
> run this:
> SELECT student.SAT, roster.pass FROM student JOIN roster ON
> student.student_id=roster.student_id
>
> Easy enough, but now I want to pass the results of this query through an
> R-script so I can perform analysis on this. How would I go about passing
the
> results of a query as a STDIN in hadoop (aka what would the syntax look
> like)?
>
> Thanks,
>


-- 
Joe Regan
Data Analyst
Chitika, Inc.

	[[alternative HTML version deleted]]