Hi All:
I am writing aggregations by building plugins with the mapred.lib.aggregate
classes to perform the bulk of the heavy lifting for a particular type of job.
I am wondering if there is some magic I can call by implementing
configure( JobConf job )
within my class (derived from ValueAggregatorBaseDescriptor) which would
allow the job to read and write compressed (zlib/gzip) data. I've seen several
references in the API to compressed data, but haven't been able to work out how
to use them within the context of "canned" map/reduces produced by the the
aggregate classes.
Any help/comments much appreciated.
Thanks very much.
C G
---------------------------------
Choose the right car based on your needs. Check out Yahoo! Autos new Car
Finder tool.
