Jason,
I'm curious to know as to how the side effect files in mapper side on job having reduce != 0 get propagated to hdfs output. As I understand, the output will be committed on the reduce side. If the side effect files from map side are…
Hi,
A couple of quick questions before going forward.
When you say chain mapreduce jobs, are you using chain-mapper APIs ? If yes, the concept of output file of one mapper as an input to other isn't that straightforward, since i believe it…
To write into multiple partitions, please look at pig's skewed join implementation of partitioner. I believe they do something pretty similar. However, .20onwards reducers will have to be set. Hence, it might break your implementation.
Coming…