Absolutely, that is one direction we are looking into, and none of us has
actively
started on that.
Bharath is planning to start on join-reordering
https://issues.apache.org/jira/browse/HIVE-1938.
Similar to the above jira, can you file one and we can discuss in more
detail in
the jira ?
Thanks,
-namit
On 2/1/11 8:19 AM, Anja Gruenheid anja.gruenh...@gatech.edu wrote:
Hi!
I'm a grad student at Georgia Tech and I'm currently working with Hive
for a university project. The project is on query optimization
techniques and possibilities in Hive. I know that there have been a lot
of additions to the ql and metastore components since the latest release
and I was hoping to help advancing those components even further. My
main interests in the course of my research is the storage and use of
metadata to run a cost-based optimizer. This involves basic
optimizations using for example the table size for cost estimations, but
also more advanced approaches using histograms. I know that table and
partition information is already collected in Hive, but from what I
could gather, column metadata and histograms are still open. Would it be
possible for me to contribute to the project in that area?
Anja
