mahout-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Dan Brickley <dan...@danbri.org>
Subject Re: Mahout Hosting Provider
Date Fri, 17 Feb 2012 04:13:52 GMT
On 17 February 2012 01:19, Lance Norskog <goksron@gmail.com> wrote:
> Is it possible to deploy Mahout over the Elastic Map/Reduce service in Amazon?

According to the Mahout Wiki, yes -
https://cwiki.apache.org/MAHOUT/mahout-on-elastic-mapreduce.html

Danny Bickson posted a detailed multipart howto a year ago,
http://bickson.blogspot.com/2011/01/how-to-install-mahout-on-amazon-ec2.html
...hopefully that is still mostly accurate. See also Grant Ingersol's
recent article at
http://www.ibm.com/developerworks/java/library/j-mahout-scaling/ walks
through using an EC2 setup with Mahout.

Nearby in the Web: http://aws.amazon.com/publicdatasets/ which
includes the data from this article (i.e.
https://cwiki.apache.org/MAHOUT/asfemail.html ). Also Google Books
ngrams data is up there at
http://lucene.472066.n3.nabble.com/mapreduce-and-google-books-n-grams-td2139194.html
and the CommonCrawl.org project's datasets, see
http://commoncrawl.org/data/accessing-the-data/ ...more walk-thru docs
on using Mahout with these last two would be great.

cheers,

Dan

Mime
View raw message