hadoop - R Reducer is not working properly in Amazon EMR -

i have done map reduce code in r run in amazon emr.

my input file format: url1 word1 word2 word3 url2 word4 word2 word3 url3 word1 word7 word2

i'm expecting output as: urls concat spaces word1 url1 url3 word2 url1 url2 url3 word3 url1 url2 .. ... ..

but emr using 3 reducers , creating 3 output files. file wise output correct, combining values, no duplicate keys. if see 3 files together, there duplicate keys.

output file 1: word1 url1 url3 word2 url1 .. ..

output file 2: word2 url2 url3 word3 url1 .. ..

see, word2 distributed 2 files. need 1 key in 1 file.

i'm using, hadoop streaming in emr. please suggest me correct settings remove duplicate keys in different files.

i assume mapper working fine. reducer:

process <- function(mat){  rows = nrow(mat) cols = ncol(mat)  for(i in 1:rows) {      for(j in i+1:rows)     {         if(j<=rows)         {             if(tostring(mat[i,1])==tostring(mat[j,1]))             {             x<-paste(mat[i,2],mat[j,2],sep=" ")             mat[i,2]=x             mat<-mat[-j,]             rows<-rows-1             }         }     } }  write.table(mat, file=stdout(), quote=false, row.names=false, col.names=false) }  reduce <- function(input){   #create column names make easier work data set   names <- c("word", "value")   cols = as.list(vector(length=2, mode="character"))   names(cols) <- names    #read input   hstablereader(file=input, cols, ignorekey=true, chunksize=100000, fun=process, sep=" ")   } 

have tried using combiner gather same keys same reducer? way should able gather words similar key single reducer. check wordcount examples combiner understand how combiner class works.