lementation[1].
>
> [1].
> https://github.com/apache/flink/blob/master/flink-clients/src/main/java/org/apache/flink/client/cli/ProgramOptions.java#L181
>
>
> Best,
> Yang
>
> Andrey Bulgakov 于2022年2月19日周六 08:30写道：
>
>> Hi Austin,
>>
> Best,
> Austin
>
> [1]:
> https://nightlies.apache.org/flink/flink-docs-release-1.14/docs/deployment/resource-providers/native_kubernetes/#configuring-flink-on-kubernetes
>
> On Fri, Feb 18, 2022 at 5:14 PM Andrey Bulgakov wrote:
>
>> Hey all,
>>
>> I&#x
r am I doing something wrong?
For context, the savepoint is produced by Flink 1.8.2 and the version I'm
trying to run on K8S is 1.14.3.
--
With regards,
Andrey Bulgakov
I guess there's no point in making it a KeyedProcessFunction since it's not
going to have access to context, timers or anything like that. So it can be
a simple InputFormat returning a DataSet of key and value tuples.
On Wed, Mar 17, 2021 at 8:37 AM Andrey Bulgakov wrote:
> Hi
is correct; basically
> it
> tries to restore the state backends as is.
>
> In you current implementation, do you see it worthwhile to try to add this?
>
> Cheers,
> Gordon
>
>
>
> --
> Sent from:
> http://apache-flink-user-mailing-list-archive.2336050.n4.nabble.com/
>
--
With regards,
Andrey Bulgakov
KeyGroupsStateHandle
objects and built an InputFormat heavily based on the code I found
in RocksDBFullRestoreOperation.java.
It ended up working extremely quickly while keeping memory and CPU usage at
the minimum.
On Tue, Mar 9, 2021 at 1:51 PM Andrey Bulgakov wrote:
> Hi all,
>
> I'm
es available to each task
manager
- enabling object reuse and modifying the tuple mapper to avoid extra tuple
allocations
- manipulating memory ratios to allocate more memory to be used as heap,
managed
- allocating 20% of memory for JVM overhead
- switching to G1GC garbage collector
Again, would appreciate any help with this.
--
With regards,
Andrey Bulgakov
