Re: Vote on Dynamic resource allocation for structured streaming [SPARK-24815]

[Adam Hobbs]

+1

From: Pavan Kotikalapudi 
Sent: Thursday, January 18, 2024 4:19:32 AM
To: Spark dev list 
Subject: Re: Vote on Dynamic resource allocation for structured streaming 
[SPARK-24815]


CAUTION: This email originated from outside of the organisation. Do not click 
links or open attachments unless you recognise the sender's full email address 
and know the content is safe.


Thanks for proposing and voting for the feature Mich.

adding some references to the thread.

  *   Jira ticket - 
SPARK-24815
  *   Design 
Doc
  *   discussion 
thread
  *   PR with initial implementation - 
https://github.com/apache/spark/pull/42352

Please vote with:

[ ] +1: Accept the proposal and start with the development.
[ ] +0
[ ] -1: I don’t think this is a good idea because …

Thank you,

Pavan

On Wed, Jan 17, 2024 at 9:52 PM Mich Talebzadeh 
mailto:mich.talebza...@gmail.com>> wrote:

+1 for me  (non binding)



Disclaimer: Use it at your own risk. Any and all responsibility for any loss, 
damage or destruction of data or any other property which may arise from 
relying on this email's technical content is explicitly disclaimed. The author 
will in no case be liable for any monetary damages arising from such loss, 
damage or destruction.





This communication is intended only for use of the addressee and may contain 
legally privileged and confidential information.
If you are not the addressee or intended recipient, you are notified that any 
dissemination, copying or use of any of the information is unauthorised.

The legal privilege and confidentiality attached to this e-mail is not waived, 
lost or destroyed by reason of a mistaken delivery to you.
If you have received this message in error, we would appreciate an immediate 
notification via e-mail to contac...@bendigoadelaide.com.au or by phoning 1300 
BENDIGO (1300 236 344), and ask that the e-mail be permanently deleted from 
your system.

Bendigo and Adelaide Bank Limited ABN 11 068 049 178

Re: Vote on Dynamic resource allocation for structured streaming [SPARK-24815]

[Pavan Kotikalapudi]

Thanks for proposing and voting for the feature Mich.

adding some references to the thread.

   - Jira ticket - SPARK-24815
   
   - Design Doc
   


   - discussion thread
   
   - PR with initial implementation -
   https://github.com/apache/spark/pull/42352

Please vote with:

[ ] +1: Accept the proposal and start with the development.
[ ] +0
[ ] -1: I don’t think this is a good idea because …

Thank you,

Pavan

On Wed, Jan 17, 2024 at 9:52 PM Mich Talebzadeh 
wrote:

>
> +1 for me  (non binding)
>
>
>
> *Disclaimer:* Use it at your own risk. Any and all responsibility for any
> loss, damage or destruction of data or any other property which may arise
> from relying on this email's technical content is explicitly disclaimed.
> The author will in no case be liable for any monetary damages arising from
> such loss, damage or destruction.
>
>
>

Vote on Dynamic resource allocation for structured streaming [SPARK-24815]

[Mich Talebzadeh]

+1 for me  (non binding)



*Disclaimer:* Use it at your own risk. Any and all responsibility for any
loss, damage or destruction of data or any other property which may arise
from relying on this email's technical content is explicitly disclaimed.
The author will in no case be liable for any monetary damages arising from
such loss, damage or destruction.

Re: [Spark-Core] Improving Reliability of spark when Executors OOM

[Mridul Muralidharan]

Hi,

  We are internally exploring adding support for dynamically changing the
resource profile of a stage based on runtime characteristics.
This includes failures due to OOM and the like, slowness due to excessive
GC, resource wastage due to excessive overprovisioning, etc.
Essentially handles scale up and scale down of resources.
Instead of baking these into the scheduler directly (which is already
complex), we are modeling it as a plugin - so that the 'business logic' of
how to handle task events and mutate state is pluggable.

The main limitation I find with mutating only the cores is the limits it
places on what kind of problems can be solved with it - and mutating
resource profiles is a much more natural way to handle this
(spark.task.cpus predates RP).

Regards,
Mridul

On Wed, Jan 17, 2024 at 9:18 AM Tom Graves 
wrote:

> It is interesting. I think there are definitely some discussion points
> around this.  reliability vs performance is always a trade off and its
> great it doesn't fail but if it doesn't meet someone's SLA now that could
> be as bad if its hard to figure out why.   I think if something like this
> kicks in, it needs to be very obvious to the user so they can see that it
> occurred.  Do you have something in place on UI or something that indicates
> this? The nice thing is also you aren't wasting memory by increasing it for
> all tasks when maybe you only need it for one or two.  The downside is you
> are only finding out after failure.
>
> I do also worry a little bit that in your blog post, the error you pointed
> out isn't a java OOM but an off heap memory issue (overhead + heap usage).
> You don't really address heap memory vs off heap in that article.  Only
> thing I see mentioned is spark.executor.memory which is heap memory.
> Obviously adjusting to only run one task is going to give that task more
> overall memory but the reasons its running out in the first place could be
> different.  If it was on heap memory for instance with more tasks I would
> expect to see more GC and not executor OOM.  If you are getting executor
> OOM you are likely using more off heap memory/stack space, etc then you
> allocated.   Ultimately it would be nice to know why that is happening and
> see if we can address it to not fail in the first place.  That could be
> extremely difficult though, especially if using software outside Spark that
> is using that memory.
>
> As Holden said,  we need to make sure this would play nice with the
> resource profiles, or potentially if we can use the resource profile
> functionality.  Theoretically you could extend this to try to get new
> executor if using dynamic allocation for instance.
>
> I agree doing a SPIP would be a good place to start to have more
> discussions.
>
> Tom
>
> On Wednesday, January 17, 2024 at 12:47:51 AM CST, kalyan <
> justfors...@gmail.com> wrote:
>
>
> Hello All,
>
> At Uber, we had recently, done some work on improving the reliability of
> spark applications in scenarios of fatter executors going out of memory and
> leading to application failure. Fatter executors are those that have more
> than 1 task running on it at a given time concurrently. This has
> significantly improved the reliability of many spark applications for us at
> Uber. We made a blog about this recently. Link:
> https://www.uber.com/en-US/blog/dynamic-executor-core-resizing-in-spark/
>
> At a high level, we have done the below changes:
>
>1. When a Task fails with the OOM of an executor, we update the core
>requirements of the task to max executor cores.
>2. When the task is picked for rescheduling, the new attempt of the
>task happens to be on an executor where no other task can run concurrently.
>All cores get allocated to this task itself.
>3. This way we ensure that the configured memory is completely at the
>disposal of a single task. Thus eliminating contention of memory.
>
> The best part of this solution is that it's reactive. It kicks in only
> when the executors fail with the OOM exception.
>
> We understand that the problem statement is very common and we expect our
> solution to be effective in many cases.
>
> There could be more cases that can be covered. Executor failing with OOM
> is like a hard signal. The framework(making the driver aware of
> what's happening with the executor) can be extended to handle scenarios of
> other forms of memory pressure like excessive spilling to disk, etc.
>
> While we had developed this on Spark 2.4.3 in-house, we would like to
> collaborate and contribute this work to the latest versions of Spark.
>
> What is the best way forward here? Will an SPIP proposal to detail the
> changes help?
>
> Regards,
> Kalyan.
> Uber India.
>

Re: [Spark-Core] Improving Reliability of spark when Executors OOM

[Tom Graves]

 It is interesting. I think there are definitely some discussion points around 
this.  reliability vs performance is always a trade off and its great it 
doesn't fail but if it doesn't meet someone's SLA now that could be as bad if 
its hard to figure out why.   I think if something like this kicks in, it needs 
to be very obvious to the user so they can see that it occurred.  Do you have 
something in place on UI or something that indicates this? The nice thing is 
also you aren't wasting memory by increasing it for all tasks when maybe you 
only need it for one or two.  The downside is you are only finding out after 
failure.
I do also worry a little bit that in your blog post, the error you pointed out 
isn't a java OOM but an off heap memory issue (overhead + heap usage).  You 
don't really address heap memory vs off heap in that article.  Only thing I see 
mentioned is spark.executor.memory which is heap memory.  Obviously adjusting 
to only run one task is going to give that task more overall memory but the 
reasons its running out in the first place could be different.  If it was on 
heap memory for instance with more tasks I would expect to see more GC and not 
executor OOM.  If you are getting executor OOM you are likely using more off 
heap memory/stack space, etc then you allocated.   Ultimately it would be nice 
to know why that is happening and see if we can address it to not fail in the 
first place.  That could be extremely difficult though, especially if using 
software outside Spark that is using that memory.
As Holden said,  we need to make sure this would play nice with the resource 
profiles, or potentially if we can use the resource profile functionality.  
Theoretically you could extend this to try to get new executor if using dynamic 
allocation for instance.  

I agree doing a SPIP would be a good place to start to have more discussions.
Tom
On Wednesday, January 17, 2024 at 12:47:51 AM CST, kalyan 
 wrote:  
 
 Hello All,
At Uber, we had recently, done some work on improving the reliability of spark 
applications in scenarios of fatter executors going out of memory and leading 
to application failure. Fatter executors are those that have more than 1 task 
running on it at a given time concurrently. This has significantly improved the 
reliability of many spark applications for us at Uber. We made a blog about 
this recently. Link: 
https://www.uber.com/en-US/blog/dynamic-executor-core-resizing-in-spark/
At a high level, we have done the below changes:   
   - When a Task fails with the OOM of an executor, we update the core 
requirements of the task to max executor cores. 
   - When the task is picked for rescheduling, the new attempt of the task 
happens to be on an executor where no other task can run concurrently. All 
cores get allocated to this task itself.
   - This way we ensure that the configured memory is completely at the 
disposal of a single task. Thus eliminating contention of memory.
The best part of this solution is that it's reactive. It kicks in only when the 
executors fail with the OOM exception.
We understand that the problem statement is very common and we expect our 
solution to be effective in many cases. There could be more cases that can be 
covered. Executor failing with OOM is like a hard signal. The framework(making 
the driver aware of what's happening with the executor) can be extended to 
handle scenarios of other forms of memory pressure like excessive spilling to 
disk, etc. 
While we had developed this on Spark 2.4.3 in-house, we would like to 
collaborate and contribute this work to the latest versions of Spark.
What is the best way forward here? Will an SPIP proposal to detail the changes 
help?
Regards,Kalyan.Uber India.  

Re: Dynamic resource allocation for structured streaming [SPARK-24815]

[Pavan Kotikalapudi]

Thanks for the +1, I will propose voting in a new thread now.

- Pavan

On Wed, Jan 17, 2024 at 5:28 PM Mich Talebzadeh 
wrote:

> I think we have discussed this enough and I consider it as a useful
> feature.. I propose a vote on it.
>
> + 1 for me
>
> Mich Talebzadeh,
> Dad | Technologist | Solutions Architect | Engineer
> London
> United Kingdom
>
>
>view my Linkedin profile
> 
>
>
>  https://en.everybodywiki.com/Mich_Talebzadeh
> 
>
>
>
> *Disclaimer:* Use it at your own risk. Any and all responsibility for any
> loss, damage or destruction of data or any other property which may arise
> from relying on this email's technical content is explicitly disclaimed.
> The author will in no case be liable for any monetary damages arising from
> such loss, damage or destruction.
>
>
>
>
> On Tue, 8 Aug 2023 at 01:30, Pavan Kotikalapudi
>  wrote:
>
>> Hi Spark Dev,
>>
>> I have extended traditional DRA to work for structured streaming
>> use-case.
>>
>> Here is an initial Implementation draft PR
>> https://github.com/apache/spark/pull/42352
>> 
>>  and
>> design doc:
>> https://docs.google.com/document/d/1_YmfCsQQb9XhRdKh0ijbc-j8JKGtGBxYsk_30NVSTWo/edit?usp=sharing
>> 
>>
>> Please review and let me know what you think.
>>
>> Thank you,
>>
>> Pavan
>>
>

Re: Dynamic resource allocation for structured streaming [SPARK-24815]

[Mich Talebzadeh]

I think we have discussed this enough and I consider it as a useful
feature.. I propose a vote on it.

+ 1 for me

Mich Talebzadeh,
Dad | Technologist | Solutions Architect | Engineer
London
United Kingdom


   view my Linkedin profile



 https://en.everybodywiki.com/Mich_Talebzadeh



*Disclaimer:* Use it at your own risk. Any and all responsibility for any
loss, damage or destruction of data or any other property which may arise
from relying on this email's technical content is explicitly disclaimed.
The author will in no case be liable for any monetary damages arising from
such loss, damage or destruction.




On Tue, 8 Aug 2023 at 01:30, Pavan Kotikalapudi
 wrote:

> Hi Spark Dev,
>
> I have extended traditional DRA to work for structured streaming
> use-case.
>
> Here is an initial Implementation draft PR
> https://github.com/apache/spark/pull/42352 and design doc:
> https://docs.google.com/document/d/1_YmfCsQQb9XhRdKh0ijbc-j8JKGtGBxYsk_30NVSTWo/edit?usp=sharing
>
> Please review and let me know what you think.
>
> Thank you,
>
> Pavan
>