yes i have joined u...@hadoop.apache.org mailing list. i think it is not possible to do in user code(map or reduce function) rather i dont want to make my changes visible to mapreduce programmer thats why i thought it is good idea to ask question here in this mailing list. On Tue, Apr 2, 2013 at 2:29 AM, Steve Loughran <ste...@hortonworks.com>wrote:
> have you considered joining the u...@hadoop.apache.org and asking the > question there? > > On 1 April 2013 17:38, Vikas Jadhav <vikascjadha...@gmail.com> wrote: > > > Hi > > > > I want process/store all data pertaining to one reducer. > > > > i want store it in some data structure depending on key for example > > > > (0,ABC) > > (0,TER) > > (1,DEF) > > (1,XYZ) > > > > for key 0 and 1 data will be in different data structure. > > > > then perform cross product of above datasets > > > > Currently i am looking in ReduceTask.java > > > > Thank You. > > > > > > > > > > On Mon, Apr 1, 2013 at 7:21 PM, Harsh J <ha...@cloudera.com> wrote: > > > > > The question should be more specific here: Do you want to process a > > > map's sorted total output or do you want to pre-process a whole > > > partition (i.e. all data pertaining to one reducer)? Former would be > > > more ideal inside MapTask.java, latter in ReduceTask.java. > > > > > > On Mon, Apr 1, 2013 at 5:36 PM, Vikas Jadhav <vikascjadha...@gmail.com > > > > > wrote: > > > > Hello > > > > > > > > I want to process output of mapper to processed before it is sent to > > > > reducer. > > > > > > > > @ what point i should hook in my code processing > > > > > > > > > > > > i guess it is ReduceTask.java file > > > > > > > > if anyone knows reagarding this please help me in this. > > > > > > > > > > > > Thank You. > > > > > > > > > > > > -- > > > > * > > > > * > > > > * > > > > > > > > Thanx and Regards* > > > > * Vikas Jadhav* > > > > > > > > > > > > -- > > > Harsh J > > > > > > > > > > > -- > > * > > * > > * > > > > Thanx and Regards* > > * Vikas Jadhav* > > > -- * * * Thanx and Regards* * Vikas Jadhav*
