Spark pools support on Yarn

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

Spark pools support on Yarn

Anton Puzanov
Hi everyone,

Spark supports in application, job concurrency execution by using pools and Spark's Fair scheduler (different than Yarn's Fair scheduler).

Is this feature supported when Yarn is used as a cluster manager? Are there special configurations I have to set or common pitfalls I need to be aware of?

Thanks,
Anton

Reply | Threaded
Open this post in threaded view
|

Re: Spark pools support on Yarn

Prabhu Josephraj
Hi Anton,

       Spark Pools / Spark Fair Scheduler is scheduling the tasks within a Spark Job. Each Spark job will have multiple stages and each stage will have multiple tasks.
This is different from YARN Fair Scheduler which schedules the jobs submitted to YARN Cluster. Spark Pools within a Spark Job will work on YARN Cluster as well.

Thanks,
Prabhu Joseph

On Tue, Feb 26, 2019 at 11:53 AM Anton Puzanov <[hidden email]> wrote:
Hi everyone,

Spark supports in application, job concurrency execution by using pools and Spark's Fair scheduler (different than Yarn's Fair scheduler).

Is this feature supported when Yarn is used as a cluster manager? Are there special configurations I have to set or common pitfalls I need to be aware of?

Thanks,
Anton