Partitions and the number of cores/executors

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

Partitions and the number of cores/executors

mbilalce.dev@gmail.com
Hi,

I am working with Gelly graph library but I think the question is applicable in general. I just want to confirm if a single data partition in Flink is executed by only a single executor/core? i.e. multiple executors can't be utilized to process a single partition in parallel. So, if I need to have utilize higher parallelism I should simply have an equal number of partitions.

I suppose the reason for this is to avoid any kind of locking that might be required to process a data partition with multiple executors.


Thanks

- Bilal

Reply | Threaded
Open this post in threaded view
|

Re: Partitions and the number of cores/executors

Stefan Richter-4
Hi,

Your assumption is right. Parallel processing is based in splitting inputs and each split is only processed by one task instance at a time.

Best,
Stefan

> On 13. Mar 2019, at 09:52, [hidden email] wrote:
>
> Hi,
>
> I am working with Gelly graph library but I think the question is applicable in general. I just want to confirm if a single data partition in Flink is executed by only a single executor/core? i.e. multiple executors can't be utilized to process a single partition in parallel. So, if I need to have utilize higher parallelism I should simply have an equal number of partitions.
>
> I suppose the reason for this is to avoid any kind of locking that might be required to process a data partition with multiple executors.
>
>
> Thanks
>
> - Bilal
>