So basically if I don't have the data into the HDFS sistem, the map reduce from HADOOP will not help me ? Because I need to build a tool that processes big data files, a lot of messages from queues or databases, and I thought that by using the map reduce from HADOOP my life would be easier :)
Thanks for the answers ________________________________ From: real great.. <greatness.hardn...@gmail.com> To: mapreduce-user@hadoop.apache.org Sent: Friday, January 27, 2012 5:15 PM Subject: Re: Question about mapReduce. I think his question was more different.Not sure though. On Fri, Jan 27, 2012 at 5:21 PM, Ashwanth Kumar <ashwanthku...@googlemail.com> wrote: Sorry Harsh, it was quite some time since I followed Sqoop. Thanks for the update. > >- Ashwanth > > >On Fri, Jan 27, 2012 at 5:07 PM, Harsh J <ha...@cloudera.com> wrote: > >Small correction to Ashwanth's post - Sqoop is now an Apache Incubator >>project residing at http://incubator.apache.org/sqoop with a community >>of its own. >> >> >>On Fri, Jan 27, 2012 at 4:56 PM, Ashwanth Kumar >><ashwanthku...@googlemail.com> wrote: >>> Hadoop is very good at processing data from HDFS. Tools like Sqoop (from >>> Cloudera) imports data from SQL databases to HDFS for processing. Everything >>> that is processed in Hadoop is only from HDFS. So if you can put anything in >>> a HDFS, you can process anything. >>> >>> - Ashwanth >>> >>> >>> On Fri, Jan 27, 2012 at 4:28 PM, neo21 zerro <neo21_ze...@yahoo.com> wrote: >>>> >>>> >>>> Hello, >>>> >>>> >>>> I'm really new to Hadoop and I was wondering if the MAP reduce >>>> programming model from Hadoop is a good choice only for processing large >>>> amount of data, from a file, database or a queue? Thanks! >>> >>> >> >> >> >>-- >>Harsh J >>Customer Ops. Engineer, Cloudera >> > -- Regards, R.V.
