Hi,
     If in a  UDF , say in the constructor of the class, i initialize a list
(say  ArrayList<String> namesList) of objects(say names). And in the exec()
method , I do some processing. When I am using this udf in a 20 node hadoop
cluster, will this list 'nameList' be instantiated multiple times or will
their be a single instance of namesList and all the mapper/reducers will use
this single one?

Thanks,
Souri

Reply via email to