Most analytic queries do require some custom code, in most cases this need is expressed in some sort of "stored procedure" language. I don't see any problem in writing these SPs/UDFs in any language that can be compiled into native code and then executing it on the Drill backend.
On Sat, Sep 15, 2012 at 3:43 AM, moon soo Lee <[email protected]> wrote: > In my understanding, user write some query(DrQL) and drill runs it using > it's distributed execution engine. > > Hive, Compare to drill, > Hive runs HiveQL using map-reduce framework. > It allow UDF and custom Map-Reduce script, and they deployed to all cluster > at runtime. > > I read proposal and saw many pluggable components. eg. query interface, > query execution engine operators. > > Will drill support user extension of it's query language? (like hive UDF) > or more? (like hive custom map-reduce script) > then how drill accomplish this? > > Thanks. >
