[jira] [Commented] (BEAM-3737) Key-aware batching function
[
https://issues.apache.org/jira/browse/BEAM-3737?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16507150#comment-16507150
]
Debasish Das commented on BEAM-3737:
I saw this is being mentioned in TFMA...I am also not clear why BatchElements()
is neededgroupByKey takes combiner which should run on both map and reduce
side...Am I missing something here ? Is it the case that beam Combiner does not
run on map side ? [~robertwb] is that why you mentioned that we should run the
combiner upfront in ParDo and then run groupByKey to achieve map and reduce
side combine ?
> Key-aware batching function
> ---
>
> Key: BEAM-3737
> URL: https://issues.apache.org/jira/browse/BEAM-3737
> Project: Beam
> Issue Type: New Feature
> Components: sdk-py-core
>Reporter: Chuan Yu Foo
>Priority: Major
>
> I have a CombineFn for which add_input has very large overhead. I would like
> to batch the incoming elements into a large batch before each call to
> add_input to reduce this overhead. In other words, I would like to do
> something like:
> {{elements | GroupByKey() | BatchElements() | CombineValues(MyCombineFn())}}
> Unfortunately, BatchElements is not key-aware, and can't be used after a
> GroupByKey to batch elements per key. I'm working around this by doing the
> batching within CombineValues, which makes the CombineFn rather messy. It
> would be nice if there were a key-aware BatchElements transform which could
> be used in this context.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
[jira] [Commented] (BEAM-3737) Key-aware batching function
[
https://issues.apache.org/jira/browse/BEAM-3737?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16426269#comment-16426269
]
Robert Bradshaw commented on BEAM-3737:
---
It seems that GroupByKey() would already give you values batched per key,
right? Or are you looking for something you can place before the GBK that
enables combiner lifting?
> Key-aware batching function
> ---
>
> Key: BEAM-3737
> URL: https://issues.apache.org/jira/browse/BEAM-3737
> Project: Beam
> Issue Type: New Feature
> Components: sdk-py-core
>Reporter: Chuan Yu Foo
>Priority: Major
>
> I have a CombineFn for which add_input has very large overhead. I would like
> to batch the incoming elements into a large batch before each call to
> add_input to reduce this overhead. In other words, I would like to do
> something like:
> {{elements | GroupByKey() | BatchElements() | CombineValues(MyCombineFn())}}
> Unfortunately, BatchElements is not key-aware, and can't be used after a
> GroupByKey to batch elements per key. I'm working around this by doing the
> batching within CombineValues, which makes the CombineFn rather messy. It
> would be nice if there were a key-aware BatchElements transform which could
> be used in this context.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
[jira] [Commented] (BEAM-3737) Key-aware batching function
[
https://issues.apache.org/jira/browse/BEAM-3737?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16387286#comment-16387286
]
Kenneth Knowles commented on BEAM-3737:
---
Actually I may have misinterpreted this, but I think [~robertwb] has context
for this. Unassigning for now.
> Key-aware batching function
> ---
>
> Key: BEAM-3737
> URL: https://issues.apache.org/jira/browse/BEAM-3737
> Project: Beam
> Issue Type: New Feature
> Components: sdk-py-core
>Reporter: Chuan Yu Foo
>Priority: Major
>
> I have a CombineFn for which add_input has very large overhead. I would like
> to batch the incoming elements into a large batch before each call to
> add_input to reduce this overhead. In other words, I would like to do
> something like:
> {{elements | GroupByKey() | BatchElements() | CombineValues(MyCombineFn())}}
> Unfortunately, BatchElements is not key-aware, and can't be used after a
> GroupByKey to batch elements per key. I'm working around this by doing the
> batching within CombineValues, which makes the CombineFn rather messy. It
> would be nice if there were a key-aware BatchElements transform which could
> be used in this context.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
[jira] [Commented] (BEAM-3737) Key-aware batching function
[
https://issues.apache.org/jira/browse/BEAM-3737?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16373867#comment-16373867
]
Kenneth Knowles commented on BEAM-3737:
---
Would you be interested in contributing something here?
> Key-aware batching function
> ---
>
> Key: BEAM-3737
> URL: https://issues.apache.org/jira/browse/BEAM-3737
> Project: Beam
> Issue Type: New Feature
> Components: beam-model
>Reporter: Chuan Yu Foo
>Assignee: Kenneth Knowles
>Priority: Major
>
> I have a CombineFn for which add_input has very large overhead. I would like
> to batch the incoming elements into a large batch before each call to
> add_input to reduce this overhead. In other words, I would like to do
> something like:
> {{elements | GroupByKey() | BatchElements() | CombineValues(MyCombineFn())}}
> Unfortunately, BatchElements is not key-aware, and can't be used after a
> GroupByKey to batch elements per key. I'm working around this by doing the
> batching within CombineValues, which makes the CombineFn rather messy. It
> would be nice if there were a key-aware BatchElements transform which could
> be used in this context.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
[jira] [Commented] (BEAM-3737) Key-aware batching function
[
https://issues.apache.org/jira/browse/BEAM-3737?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16373866#comment-16373866
]
Kenneth Knowles commented on BEAM-3737:
---
This is a very interesting scenario!
Can you build a {{BatchingCombineFn}} that has an accumulator that just buffers
elements for a while and compacts them as needed? Then in {{extractOutput}} you
can do the final conversion.
> Key-aware batching function
> ---
>
> Key: BEAM-3737
> URL: https://issues.apache.org/jira/browse/BEAM-3737
> Project: Beam
> Issue Type: New Feature
> Components: beam-model
>Reporter: Chuan Yu Foo
>Assignee: Kenneth Knowles
>Priority: Major
>
> I have a CombineFn for which add_input has very large overhead. I would like
> to batch the incoming elements into a large batch before each call to
> add_input to reduce this overhead. In other words, I would like to do
> something like:
> {{elements | GroupByKey() | BatchElements() | CombineValues(MyCombineFn())}}
> Unfortunately, BatchElements is not key-aware, and can't be used after a
> GroupByKey to batch elements per key. I'm working around this by doing the
> batching within CombineValues, which makes the CombineFn rather messy. It
> would be nice if there were a key-aware BatchElements transform which could
> be used in this context.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
