I had the same choices to do ...
The result are : whenever I can work with the new API, I do
if not, I continue working with the OLD API (as for MultipleOutputFormat)
The whole process is thus a mix of new API jobs mixed with old API jobs
: works perfectly.
MultipleOutputFormat is out with the new API with the 0.21 hadoop version :)
Hope it helps
Olivier Varene
Le 19/11/2010 00:15, Srihari Anantha Padmanabhan a écrit :
Hi,
I am working on migrating a mapreduce program from using org.apache.hadoop.mapred to org.apache.hadoop.mapreduce APIs. The program currently uses orf.apache.hadoop.mapred.lib.MultipleOutputFormat<k,v>. I could not find any equivalent class in mapreduce.
Can anyone suggest an equivalent class or a workaround for using a mapreduce API instead of mapred.lib.MultipleOutputFormat<k,v>?
Will there be any issue if I continue to use mapred?
Any reference on this would be very helpful.
Thanks,
Srihari
I am working on migrating a mapreduce program from using org.apache.hadoop.mapred to org.apache.hadoop.mapreduce APIs. The program currently uses orf.apache.hadoop.mapred.lib.MultipleOutputFormat<k,v>. I could not find any equivalent class in mapreduce.
Can anyone suggest an equivalent class or a workaround for using a mapreduce API instead of mapred.lib.MultipleOutputFormat<k,v>?
Will there be any issue if I continue to use mapred?
Any reference on this would be very helpful.
Thanks,
Srihari
