[GitHub] spark pull request: [SPARK-5037][STREAMING] dynamically loaded DSt...

[tdas]

Github user tdas commented on the pull request:

https://github.com/apache/spark/pull/3858#issuecomment-120189769
  
Okay. I hope you get time in future.

TD

On Thu, Jul 9, 2015 at 5:16 PM, industrial-sloth 
wrote:

> Closed #3858 .
>
> —
> Reply to this email directly or view it on GitHub
> .
>



---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org

[GitHub] spark pull request: [SPARK-5037][STREAMING] dynamically loaded DSt...

[industrial-sloth]

Github user industrial-sloth closed the pull request at:

https://github.com/apache/spark/pull/3858


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org

[GitHub] spark pull request: [SPARK-5037][STREAMING] dynamically loaded DSt...

[industrial-sloth]

Github user industrial-sloth commented on the pull request:

https://github.com/apache/spark/pull/3858#issuecomment-120179506
  
Ha, that was an old throwaway account. Anyway, as I was saying, no updates, 
and no reasonable chance of progress in the near future either. Fine by me to 
close out the PR and JIRA.


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org

[GitHub] spark pull request: [SPARK-5037][STREAMING] dynamically loaded DSt...

[SlothPetro]

Github user SlothPetro commented on the pull request:

https://github.com/apache/spark/pull/3858#issuecomment-120179245
  
Hi @tdas - Unfortunately no, and it doesn't look likely that this will be 
changing any time in the near future, either. I'd be fine with just closing out 
the PR and JIRA ticket - someone else can pick up with a clean implementation 
more specifically focused on ZMQ if there's a current need for it.


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org

[GitHub] spark pull request: [SPARK-5037][STREAMING] dynamically loaded DSt...

[tdas]

Github user tdas commented on the pull request:

https://github.com/apache/spark/pull/3858#issuecomment-120148480
  
@industrial-sloth Hey any updates on this?



---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org

[GitHub] spark pull request: [SPARK-5037][STREAMING] dynamically loaded DSt...

[zsxwing]

Github user zsxwing commented on the pull request:

https://github.com/apache/spark/pull/3858#issuecomment-114140106
  
@industrial-sloth I think most of work to add a new type of DStream 
includes:

1. Convert parameters from Python types to Java types
2. Convert DStream[T] from Python to Java (T may not have a consistent type 
in Python, see #6830 for example).
3. Add an assembly jar to include all dependencies not in spark assembly 
jar, so that the Python user can use the Python API easily. Otherwise, the user 
needs to figure out which jars should be added to `-jars`

And I don't think `reflectedStream` will reduce any work.

For the ZeroMQ Python API, you also need to add an assembly jar since 
`akka-zeromq` is not in the Spark assembly jar.


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org

[GitHub] spark pull request: [SPARK-5037][STREAMING] dynamically loaded DSt...

[tdas]

Github user tdas commented on the pull request:

https://github.com/apache/spark/pull/3858#issuecomment-113695466
  
Could you also update the JIRA and title to make it more specific to the
problem? E.g. Python API for ZeroMQ?

On Fri, Jun 19, 2015 at 3:05 PM, industrial-sloth 
wrote:

> Sounds good @tdas . I'll try to take a look and
> get this updated sometime over the next week or so. Will ping you with a
> comment here when there's something worth looking at. :)
>
> Thanks for checking this out!
>
> —
> Reply to this email directly or view it on GitHub
> .
>



---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org

[GitHub] spark pull request: [SPARK-5037][STREAMING] dynamically loaded DSt...

[industrial-sloth]

Github user industrial-sloth commented on the pull request:

https://github.com/apache/spark/pull/3858#issuecomment-113656363
  
Sounds good @tdas. I'll try to take a look and get this updated sometime 
over the next week or so. Will ping you with a comment here when there's 
something worth looking at. :)

Thanks for checking this out!


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org

[GitHub] spark pull request: [SPARK-5037][STREAMING] dynamically loaded DSt...

[tdas]

Github user tdas commented on the pull request:

https://github.com/apache/spark/pull/3858#issuecomment-113648094
  
That gives me some context. We are adding more stuff to the python API for
parity with Scala and Java. We have added full Kafka Python API, Flume is
being added, and hopefully Kinesis will also be added. The technique by
which they have been added is similar to your approach at high level, but
not quite the same. Take a look at the python KafkaUtils in the current
code. Since we want to maintain consistent in design, I am happy to take a
look if you can update the current PR to use the existing style.

On Fri, Jun 19, 2015 at 1:22 PM, industrial-sloth 
wrote:

> Sure thing @tdas . First a caveat: I haven't
> been keeping up with the spark community since ~March 2015, so the issues 
I
> originally hit might no longer exist w/ more recent spark releases.
>
> As of December 2014 we were exploring streaming options for real time
> analysis in Thunder (https://github.com/thunder-project). Thunder is
> pyspark based; at that time our pyspark dstream options, as I recall, were
> basically either file-based (watch a directory for new files) or to
> integrate with Kafka. Specifically there was no option to listen to a
> ZeroMQ stream or to many of the other dstream types available in the scala
> API. We wanted to be pushing a high-bandwidth stream of microscope images
> over to pyspark for further analysis. ZeroMQ seemed ideal; Kafka seemed
> like too much and file-based seemed to necessitate an additional
> unnecessary disk IO. So I put together a ZMQ solution for pyspark 
streaming
> and threw it out there in this PR.
>
> Again, haven't been keeping up, not sure whether this is still a concern
> w/ current releases of pyspark. I agree this is potentially an unusual use
> case - our workaround at the time was to go to the file-based dstream
> implementation, which was functional but perhaps not optimal.
>
> Any further comment on this @freeman-lab 
> or @andrewosh ?
>
> —
> Reply to this email directly or view it on GitHub
> .
>



---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org

[GitHub] spark pull request: [SPARK-5037][STREAMING] dynamically loaded DSt...

[industrial-sloth]

Github user industrial-sloth commented on the pull request:

https://github.com/apache/spark/pull/3858#issuecomment-113631381
  
Sure thing @tdas. First a caveat: I haven't been keeping up with the spark 
community since ~March 2015, so the issues I originally hit might no longer 
exist w/ more recent spark releases. 

As of December 2014 we were exploring streaming options for real time 
analysis in Thunder (https://github.com/thunder-project). Thunder is pyspark 
based; at that time our pyspark dstream options, as I recall, were basically 
either file-based (watch a directory for new files) or to integrate with Kafka. 
Specifically there was no option to listen to a ZeroMQ stream or to many of the 
other dstream types available in the scala API. We wanted to be pushing a 
high-bandwidth stream of microscope images over to pyspark for further 
analysis. ZeroMQ seemed ideal; Kafka seemed like too much and file-based seemed 
to necessitate an additional unnecessary disk IO. So I put together a ZMQ 
solution for pyspark streaming and threw it out there in this PR. 

Again, haven't been keeping up, not sure whether this is still a concern w/ 
current releases of pyspark. I agree this is potentially an unusual use case - 
our workaround at the time was to go to the file-based dstream implementation, 
which was functional but perhaps not optimal.

Any further comment on this @freeman-lab  or @andrewosh?


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org

[GitHub] spark pull request: [SPARK-5037][STREAMING] dynamically loaded DSt...

[tdas]

Github user tdas commented on the pull request:

https://github.com/apache/spark/pull/3858#issuecomment-113590414
  
The idea is interesting. But I am not clear what sort of concrete 
situations does it improve. @industrial-sloth could you provide us an example?


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org

[GitHub] spark pull request: [SPARK-5037][STREAMING] dynamically loaded DSt...

[andrewor14]

Github user andrewor14 commented on the pull request:

https://github.com/apache/spark/pull/3858#issuecomment-113321838
  
@tdas is this something we want?


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org

[GitHub] spark pull request: [SPARK-5037][STREAMING] dynamically loaded DSt...

[AmplabJenkins]

Github user AmplabJenkins commented on the pull request:

https://github.com/apache/spark/pull/3858#issuecomment-96769900
  
Can one of the admins verify this patch?


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org

[GitHub] spark pull request: [SPARK-5037][STREAMING] dynamically loaded DSt...

[industrial-sloth]

Github user industrial-sloth commented on the pull request:

https://github.com/apache/spark/pull/3858#issuecomment-78387422
  
@sryza No problem, done.


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org