This question would be more appropriate on the mapreduce-users at
hadoop.apache.org mailing list.
Unfortunately Hadoop's website is out of date right now and does not
describe the currently correct mailing lists. Is someone working to fix
this? We do not want general@ to become a place for end-user questions.
Doug
乔木 wrote:
Hi, everyone
I've been learning hadoop recently and I'm confused about the combiner
mechanism.
There is a property min.num.spills.for.combine specifying the minimum number
of spills to run combiner when merging. The default value is 3. Why there is
such a restriction? Should it be better that run the combiner no matter how
many spills there are?
The second question is why the combiner could be run at the reduce side.
Can't the reduce function take place of that?
Thanks very much.
