Saw this in the Spark dev list and thought it's appropriate here as well. Tim
---------- Forwarded message ---------- From: Lijie Xu <[email protected]> Date: Tue, Sep 3, 2013 at 7:33 PM Subject: About a paper titiled "Memory-Efficient GroupBy-Aggregate using Compressed Buffer Trees" To: [email protected] Hi, all. I noticed a paper called "Memory-Efficient GroupBy-Aggregate using Compressed Buffer Trees" in SoCC 2013 accepted list yesterday. Check out http://www.socc2013.org/papers. The detailed paper can be found here ( http://www.cercs.gatech.edu/tech-reports/tr2012/git-cercs-12-08.pdf) and there are simplified slides ( http://www.cercs.gatech.edu/iab13/material/amur.pdf). It argues another GroupBy-Aggregate approach which is different with MapReduce's Sort and Spark's hash-based aggregatation. It seems a new way but I cannot understand the pros and cons of this approach. Any comments and discussion are welcome. -- You received this message because you are subscribed to the Google Groups "Spark Developers" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/groups/opt_out.
