AnsweredAssumed Answered

Mapper tasks and data locality?

Question asked by hastings on Feb 2, 2015
Latest reply on Feb 3, 2015 by schandhok
When I run a M/R task I see that a lot of mappers seem to get scheduled on data nodes where the data is not local. Any thing I can do to get locality 100% of the time? If not what % of the mappers do get locality in a 100 node cluster with over 200 TB of data.

Outcomes