capacity - resource based scheduling

[Corbin Hoenes]

I have a streaming job that uses a lot of memory.  Capacity scheduler
lets me set the mapred.job.map.memory.mb property to something high
like 2560.  The job then takes 5 slots (512mb slot) for every map
task.  I have noticed that it appears to actually start many java
processes that look like tasks (I've seen as many as 12) on my slaves
to service my map tasks.  What are all the extra java processes for
(are these considered slots?) they use memory but no cpu, are they
just place holders?