This is an automatically generated e-mail. To reply, visit:


    In this if-branch, is it not possible for currOp to also end in a POLocalRearrange, i.e. don't you want to check if you should set prevOp=currOp?
    In the case of Map-Reduce-Reduce (implemented as Tez vertices), would not the middle Reduce start with a POPackage but then itself end in a POLocalRearrange?
- Alex Bain
On Nov. 6, 2013, 3:55 p.m., Cheolsoo Park wrote: