Support for Auto scaling

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

Support for Auto scaling

sandeep6
Hi Guys,

I currently run flink 1.1.4 streaming jobs in EMR in AWS with yarn. I understand that EMR supports auto scaling but Flink doesn't. Is there a plan for this support in 1.2. 

Thanks,
Sandeep
Reply | Threaded
Open this post in threaded view
|

Re: Support for Auto scaling

Tzu-Li (Gordon) Tai
Hi Sandeep!

While auto scaling jobs in Flink still isn’t possible, in Flink 1.2 you will be able to rescale jobs by stopping and restarting.
This works by taking a savepoint of the job before stopping the job, and then redeploy the job with a higher / lower parallelism using the savepoint.
Upon restarting the job, your states will be redistributed across the new operators.

Changing operator / job parallelism on the fly while running is still on the future roadmap.

Cheers,
Gordon

On February 2, 2017 at 8:39:39 AM, Meghashyam Sandeep V ([hidden email]) wrote:

Hi Guys,

I currently run flink 1.1.4 streaming jobs in EMR in AWS with yarn. I understand that EMR supports auto scaling but Flink doesn't. Is there a plan for this support in 1.2. 

Thanks,
Sandeep
Reply | Threaded
Open this post in threaded view
|

Re: Support for Auto scaling

Till Rohrmann
Hi Sandeep,

it's also not yet supported to dynamically start new TaskManagers when needed. However, it can be done manually via the taskmanager.sh script.

We're working to enable this feature for 1.3 if everything goes according to plan. 

Cheers,
Till

On Thu, Feb 2, 2017 at 6:49 AM, Tzu-Li (Gordon) Tai <[hidden email]> wrote:
Hi Sandeep!

While auto scaling jobs in Flink still isn’t possible, in Flink 1.2 you will be able to rescale jobs by stopping and restarting.
This works by taking a savepoint of the job before stopping the job, and then redeploy the job with a higher / lower parallelism using the savepoint.
Upon restarting the job, your states will be redistributed across the new operators.

Changing operator / job parallelism on the fly while running is still on the future roadmap.

Cheers,
Gordon


On February 2, 2017 at 8:39:39 AM, Meghashyam Sandeep V ([hidden email]) wrote:

Hi Guys,

I currently run flink 1.1.4 streaming jobs in EMR in AWS with yarn. I understand that EMR supports auto scaling but Flink doesn't. Is there a plan for this support in 1.2. 

Thanks,
Sandeep