This is a consequence of how map-reduce works. It does hash partitioning, but then on each reducer the keys are sorted before it begins the reducing. It does not do hash partitioning *within* reducers.
On Mon, Jan 14, 2019 at 12:49 PM Russell Carden <[email protected]> wrote: > I do a sort by grouping on a key and then doing a sortby on the values. I > then write out the results. Since I have multiple reducers, I will have > multiple files. I have observed that in the output for each reducer, the > keys are also sorted. That is not only are the values within each group > sorted, the groups themselves are sorted in each file. Is this a > consequence of how the results for each group are merged within a reducer? > > -- > You received this message because you are subscribed to the Google Groups > "Scalding Development" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > For more options, visit https://groups.google.com/d/optout. > -- You received this message because you are subscribed to the Google Groups "Scalding Development" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
