[
https://issues.apache.org/jira/browse/BEAM-2085?focusedWorklogId=585427&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-585427
]
ASF GitHub Bot logged work on BEAM-2085:
----------------------------------------
Author: ASF GitHub Bot
Created on: 19/Apr/21 22:56
Start Date: 19/Apr/21 22:56
Worklog Time Spent: 10m
Work Description: TheNeuralBit commented on a change in pull request
#14390:
URL: https://github.com/apache/beam/pull/14390#discussion_r616230075
##########
File path: sdks/python/apache_beam/runners/pipeline_context.py
##########
@@ -118,13 +120,19 @@ def get_by_id(self, id):
def get_by_proto(self, maybe_new_proto, label=None, deduplicate=False):
# type: (message.Message, Optional[str], bool) -> str
+ obj = self._obj_type.from_runner_api(
Review comment:
Hm I'll need to spend my time to jog my memory on this. I think I may
have naively propagated some existing logic, prior to #12067 the return
statement was:
```
return self.put_proto(self._unique_ref(label), maybe_new_proto)
```
And the first argument of `unique_ref` is `obj`. I vaguely recall puzzling
over this for a while.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
Issue Time Tracking
-------------------
Worklog Id: (was: 585427)
Time Spent: 22h (was: 21h 50m)
> Transforms could give hints to runners on resource requirements
> ---------------------------------------------------------------
>
> Key: BEAM-2085
> URL: https://issues.apache.org/jira/browse/BEAM-2085
> Project: Beam
> Issue Type: Improvement
> Components: beam-model, runner-core, sdk-java-core
> Affects Versions: Not applicable
> Reporter: Ismaël Mejía
> Assignee: Valentyn Tymofieiev
> Priority: P3
> Labels: Clarified
> Time Spent: 22h
> Remaining Estimate: 0h
>
> As discussed in BEAM-673 runners can allocate workers to accomplish their
> work in a better way if they can take into account some hints from the
> transforms, e.g. a source can hint data locality and with this information
> the runner can allocate the workers in an better way, this can also be the
> case with a particular transform that can suggest to the runner to be
> executed in a worker with a specific resource, e.g. GPU.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
