Re: Shuffle guarantees

[Corey Nolet]

Nevermind, a look @ the ExternalSorter class tells me that the iterator for
each key that's only partially ordered ends up being merge sorted by
equality after the fact. Wanted to post my finding on here for others who
may have the same questions.


On Tue, Mar 1, 2016 at 3:05 PM, Corey Nolet  wrote:

> The reason I'm asking, I see a comment in the ExternalSorter class that
> says this:
>
> "If we need to aggregate by key, we either use a total ordering from the
> ordering parameter, or read the keys with the same hash code and compare
> them with each other for equality to merge values".
>
> How can this be assumed if the object used for the key, for instance, in
> the case where a HashPartitioner is used, cannot assume ordering and
> therefore cannot assume a comparator can be used?
>
> On Tue, Mar 1, 2016 at 2:56 PM, Corey Nolet  wrote:
>
>> So if I'm using reduceByKey() with a HashPartitioner, I understand that
>> the hashCode() of my key is used to create the underlying shuffle files.
>>
>> Is anything other than hashCode() used in the shuffle files when the data
>> is pulled into the reducers and run through the reduce function? The reason
>> I'm asking is because there's a possibility of hashCode() colliding in two
>> different objects which end up hashing to the same number, right?
>>
>>
>>
>

Re: Shuffle guarantees

[Corey Nolet]

The reason I'm asking, I see a comment in the ExternalSorter class that
says this:

"If we need to aggregate by key, we either use a total ordering from the
ordering parameter, or read the keys with the same hash code and compare
them with each other for equality to merge values".

How can this be assumed if the object used for the key, for instance, in
the case where a HashPartitioner is used, cannot assume ordering and
therefore cannot assume a comparator can be used?

On Tue, Mar 1, 2016 at 2:56 PM, Corey Nolet  wrote:

> So if I'm using reduceByKey() with a HashPartitioner, I understand that
> the hashCode() of my key is used to create the underlying shuffle files.
>
> Is anything other than hashCode() used in the shuffle files when the data
> is pulled into the reducers and run through the reduce function? The reason
> I'm asking is because there's a possibility of hashCode() colliding in two
> different objects which end up hashing to the same number, right?
>
>
>

Shuffle guarantees

[Corey Nolet]

So if I'm using reduceByKey() with a HashPartitioner, I understand that the
hashCode() of my key is used to create the underlying shuffle files.

Is anything other than hashCode() used in the shuffle files when the data
is pulled into the reducers and run through the reduce function? The reason
I'm asking is because there's a possibility of hashCode() colliding in two
different objects which end up hashing to the same number, right?