Alan, Can such cases be found out by running through explain/illustrate? Thanks!
Michael ________________________________ From: Alan Gates <[email protected]> To: [email protected] Sent: Thursday, September 15, 2011 7:44 AM Subject: Re: Num Reducers is Very Small! Can you share your script? There are certain operations like group all that cannot be parallelized. But in general when you set parallel it should be done in parallel. Alan. On Sep 15, 2011, at 12:47 AM, 唐亮 wrote: > Hi dear pigs: > Sometimes when I run pig job with huge data, > the number of reducers is very small(only 1 reducer), even if set PARALLEL, > so the job runs extremely slow! > > Can I or how can I increase the number of reducers in pig job? > Thanks!
