+1
I’d prefer to see more of the end goal and how that could be achieved (such as
ETL or SPARK-24579). However given the rounds and months of discussions we have
come down to just the public API.
If the community thinks a new set of public API is maintainable, I don’t see
any problem with that
Sure im not talking about k8s here.
The discussion is about the heuristics and their drawbacks.
Στις Δευ, 27 Μαΐ 2019, 2:04 μ.μ. ο χρήστης Gabor Somogyi <
gabor.g.somo...@gmail.com> έγραψε:
> K8s is a different story, please take a look at the doc "Future Work" part.
>
> On Fri, May 24, 2019 at 9
Hey Jörn,
Thanks a lot for replying.
My Data Source extends BaseRelation and PrunedFilteredScan. The buildScan()
returns my custom RDD. I want to take a lock before any executors start
reading the data, and release it after all executors are done, and so I
tried to acquire lock in MyRDD.getPartit
K8s is a different story, please take a look at the doc "Future Work" part.
On Fri, May 24, 2019 at 9:40 PM Stavros Kontopoulos <
stavros.kontopou...@lightbend.com> wrote:
> Btw the heuristics for batch mode (
> https://github.com/apache/spark/blob/master/core/src/main/scala/org/apache/spark/Exec
