Thanks Sameer and Till,
On Mon, Aug 1, 2016 at 9:31 AM, Till Rohrmann wrote:
> Yes you're right Sameer. That's how things work in Flink.
>
> Cheers,
> Till
>
> On Sun, Jul 31, 2016 at 12:33 PM, Sameer Wadkar
> wrote:
>
>> Vishnu,
>>
>> I would imagine based on Max's explanation and how other s
Yes you're right Sameer. That's how things work in Flink.
Cheers,
Till
On Sun, Jul 31, 2016 at 12:33 PM, Sameer Wadkar wrote:
> Vishnu,
>
> I would imagine based on Max's explanation and how other systems like
> MapReduce and Spark partition keys, if you have 100 keys and 50 slots, 2
> keys wou
Vishnu,
I would imagine based on Max's explanation and how other systems like MapReduce
and Spark partition keys, if you have 100 keys and 50 slots, 2 keys would be
assigned to each slot. Each slot would maintain one or more windows (more for
time based windows) and each window would have upto
Hi Max,
Thanks for the explanation.
"This happens one after another in a single task slot but in parallel
across all the task slots".
Could you explain more on how this happens in parallel? Which part does
occur in parallel? Is it the Trigger going through each pane and the window
function being
Hi Vishnu Viswanath,
The keyed elements are spread across the 50 task slots (assuming you
have a parallelism of 50) using hash partitioning on the keys. Each
task slot runs one or multiple operators (depending on the slot
sharing options). One of them is a WindowOperator which will decide
when to
Hi,
Lets say I have a window on a keyed stream, and I have about 100 unique
keys.
And assume I have about 50 tasks slots in my cluster. And suppose my
trigger fired 70/100 windows/pane at the same time.
How will flink handle this? Will it assign 50/70 triggered windows to the
50 available task sl
