[
https://issues.apache.org/jira/browse/APEXCORE-714?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15985816#comment-15985816
]
Pramod Immaneni commented on APEXCORE-714:
------------------------------------------
Thomas,
My responses are inline..
On Tue, Apr 25, 2017 at 8:42 AM, Thomas Weise <t...@apache.org> wrote:
Pramod,
Sounds like some sort of alternative "processing mode" that from engine
perspective allows potentially inconsistent state when there is a pipeline
failure. This is of course only something the user can decide.
>> Calling it an alternate processing mode is a good idea.
Does the proposal assume that the operator state is immutable (or what is
sometimes tagged with the stateless annotation)? For example an operator
that has to load a large amount of state from another source before it can
process the first tuple?
>> Operator state can change, not necessarily stateless. Stateless may not
>> automatically fall into this category as our current definition of stateless
>> denotes window level stateless and not necessarily tuple level.
Also, it would be an optimization but not something that will help with SLA
if the operator still needs to be recovered when its own container fails.
It might help to clarify that and also why there is a need to recover in
the batch use case (vs. reprocess).
>> Correct, as I mentioned in the last statement, if the container where the
>> operator is running itself goes down then it is recovery from checkpoint and
>> business as usual. I may have misspoken about batch, meant to say apps where
>> operators have large state not necessarily batch, the use case we are
>> dealing with is batch and restart is not practical as the run takes a long
>> time.
Thanks,
Pramod
> Reusable instance operator recovery
> -----------------------------------
>
> Key: APEXCORE-714
> URL: https://issues.apache.org/jira/browse/APEXCORE-714
> Project: Apache Apex Core
> Issue Type: Improvement
> Reporter: Pramod Immaneni
> Assignee: Pramod Immaneni
>
> In a failure scenario, when a container fails, it is redeployed along with
> all the operators in it. The operators downstream to these operators are also
> redeployed within their containers. The operators are restored from their
> checkpoint and connect to the appropriate point in the stream according to
> the processing mode. In at least once mode, for example, the data is replayed
> from the same checkpoint
> Restoring an operator state from checkpoint could turn out to be a costly
> operation depending on the size of the state. In some use cases, based on the
> operator logic, when there is an upstream failure, without restoring the
> operator from checkpoint and reusing the current instance, will still produce
> the same results with the data replayed from the last fully processed window.
> The operator state can remain the same as it was before the upstream failure
> by reusing the same operator instance from before and only the streams and
> window reset to the window after the last fully processed window to guarantee
> the at least once processing of tuples. If the container where the operator
> itself is running goes down, it would need to be restored from the checkpoint
> of course. This scenario occurs in some batch use cases with operators that
> have a large state.
> I would like to propose adding the ability for a user to explicitly identify
> operators to be of this type and the corresponding functionality in the
> engine to handle their recovery in the way described above by not restoring
> their state from checkpoint, reusing the instance and restoring the stream to
> the window after the last fully processed window for the operator. When
> operators are not identified to be of this type, the default behavior is what
> it is today and nothing changes.
> I have done some prototyping on the engine side to ensure that this is
> possible with our current code base without requiring a massive overhaul,
> especially the restoration of the operator instance within the Node in the
> streaming container, the re-establishment of the subscriber stream to a
> window in the buffer server where the publisher (upstream) hasn't yet reached
> as it would be restarting from checkpoint and have been able to get it all
> working successfully.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
