Github user wesm commented on the issue:
https://github.com/apache/spark/pull/15821
Notable that this is the same technology that @hadley and I used to create
Feather last year (https://github.com/wesm/feather). We'll be continuing to
pursue collaborations in this area between the R a
Github user BryanCutler commented on the issue:
https://github.com/apache/spark/pull/15821
## Future Work
The support of all Spark data types to include complex types will be a
quick follow up to ensure full compatibility. With the added ability to
convert a Dataset on the e
Github user BryanCutler commented on the issue:
https://github.com/apache/spark/pull/15821
## Benchmarks with Conversion done on Executor (updated)
## 1mm Longs
_ | With Arrow | Without Arrow
--||---
count | 50.00 |50.00
Github user BryanCutler commented on the issue:
https://github.com/apache/spark/pull/15821
## Dependency Info
This change does add Apache Arrow as a dependency, specifically the Java
arrow-vector artifact. For Python, usage is optional and test are conditional
on ability to
Github user BryanCutler commented on the issue:
https://github.com/apache/spark/pull/15821
The Python Arrow tests should have been skipped since they are conditional
on pyarrow being installed and it's not in `requirements.txt`. I'll hold off
on testing for now until we can make sure
Github user wesm commented on the issue:
https://github.com/apache/spark/pull/15821
@BryanCutler not sure if it's relevant, but though the conda-forge
artifacts are not up to date, but they should still be wire compatible with the
Arrow 0.2 JAR
---
If your project is set up for it,
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/15821
Test FAILed.
Refer to this link for build results (access rights to CI server needed):
https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/73366/
Test FAILed.
---
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/15821
Merged build finished. Test FAILed.
---
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
e
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/15821
**[Test build #73366 has
finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/73366/testReport)**
for PR 15821 at commit
[`9c8ea63`](https://github.com/apache/spark/commit/9
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/15821
**[Test build #73366 has
started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/73366/testReport)**
for PR 15821 at commit
[`9c8ea63`](https://github.com/apache/spark/commit/9c
Github user BryanCutler commented on the issue:
https://github.com/apache/spark/pull/15821
jenkins retest this please
---
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
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/15821
Merged build finished. Test FAILed.
---
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
e
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/15821
Test FAILed.
Refer to this link for build results (access rights to CI server needed):
https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/73355/
Test FAILed.
---
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/15821
**[Test build #73355 has
finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/73355/testReport)**
for PR 15821 at commit
[`9c8ea63`](https://github.com/apache/spark/commit/9
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/15821
**[Test build #73355 has
started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/73355/testReport)**
for PR 15821 at commit
[`9c8ea63`](https://github.com/apache/spark/commit/9c
Github user BryanCutler commented on the issue:
https://github.com/apache/spark/pull/15821
Jenkins retest please
---
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 wishe
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/15821
Test FAILed.
Refer to this link for build results (access rights to CI server needed):
https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/73321/
Test FAILed.
---
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/15821
Merged build finished. Test FAILed.
---
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
e
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/15821
**[Test build #73321 has
finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/73321/testReport)**
for PR 15821 at commit
[`9c8ea63`](https://github.com/apache/spark/commit/9
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/15821
**[Test build #73321 has
started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/73321/testReport)**
for PR 15821 at commit
[`9c8ea63`](https://github.com/apache/spark/commit/9c
Github user BryanCutler commented on the issue:
https://github.com/apache/spark/pull/15821
jenkins retest this please
---
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
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/15821
Test FAILed.
Refer to this link for build results (access rights to CI server needed):
https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/73310/
Test FAILed.
---
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/15821
Merged build finished. Test FAILed.
---
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
e
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/15821
**[Test build #73310 has
finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/73310/testReport)**
for PR 15821 at commit
[`9c8ea63`](https://github.com/apache/spark/commit/9
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/15821
**[Test build #73310 has
started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/73310/testReport)**
for PR 15821 at commit
[`9c8ea63`](https://github.com/apache/spark/commit/9c
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/15821
**[Test build #73308 has
finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/73308/testReport)**
for PR 15821 at commit
[`42af1d5`](https://github.com/apache/spark/commit/4
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/15821
Test FAILed.
Refer to this link for build results (access rights to CI server needed):
https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/73308/
Test FAILed.
---
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/15821
Merged build finished. Test FAILed.
---
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
e
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/15821
**[Test build #73308 has
started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/73308/testReport)**
for PR 15821 at commit
[`42af1d5`](https://github.com/apache/spark/commit/42
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/15821
Test FAILed.
Refer to this link for build results (access rights to CI server needed):
https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/73305/
Test FAILed.
---
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/15821
**[Test build #73305 has
finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/73305/testReport)**
for PR 15821 at commit
[`54884ed`](https://github.com/apache/spark/commit/5
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/15821
Merged build finished. Test FAILed.
---
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
e
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/15821
**[Test build #73305 has
started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/73305/testReport)**
for PR 15821 at commit
[`54884ed`](https://github.com/apache/spark/commit/54
Github user BryanCutler commented on the issue:
https://github.com/apache/spark/pull/15821
Great news @wesm! I'm am just cleaning up the tests and will post an
update using the new release soon. I'm happy to help with any maintenance too,
just let me know what you need.
---
If you
Github user wesm commented on the issue:
https://github.com/apache/spark/pull/15821
The 0.2 Maven artifacts have been posted. I'll try to update the
conda-forge packages this week -- if anyone can help with conda-forge
maintenance that would be a big help.
Thanks!
---
If yo
Github user mariusvniekerk commented on the issue:
https://github.com/apache/spark/pull/15821
Probably a good thing to look at is the R pieces since that is effectively
constrained to InternalRow
---
If your project is set up for it, you can reply to this email and have your
reply ap
Github user wesm commented on the issue:
https://github.com/apache/spark/pull/15821
@zero323 that is exactly the plan. It's a bit complicated though because
the Python UDF code path handles arbitrary iterators, not just
`Array[InternalRow]`
---
If your project is set up for it, you
Github user zero323 commented on the issue:
https://github.com/apache/spark/pull/15821
This looks amazing and I can't help but wonder - if the next step is
_generating the arrow batches on executors_ is it possible to reuse this to
pass data between JVM and Python UDFs? Right now, wit
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/15821
Test FAILed.
Refer to this link for build results (access rights to CI server needed):
https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/72180/
Test FAILed.
---
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/15821
Merged build finished. Test FAILed.
---
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
e
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/15821
**[Test build #72180 has
started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/72180/testReport)**
for PR 15821 at commit
[`b35192c`](https://github.com/apache/spark/commit/b3
Github user wesm commented on the issue:
https://github.com/apache/spark/pull/15821
The conda-forge pyarrow package is now up to date with latest Arrow git
master
---
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 p
Github user BryanCutler commented on the issue:
https://github.com/apache/spark/pull/15821
Thanks for the review @wesm! Those are good ideas, I'll work on an update
to this. As for the packaging conflicts, I had to add exclusions for
`com.fasterxml.jackson.core:jackson-annotations`,
Github user BryanCutler commented on the issue:
https://github.com/apache/spark/pull/15821
> Parallelizing the record batch conversion and streaming it to Python
would be another significant perf win.
Right, I should have also mentioned that this PR takes a simplistic
approa
Github user wesm commented on the issue:
https://github.com/apache/spark/pull/15821
Very nice to see the improved wall clock times. I have been busy
engineering the pipeline between the byte stream from Spark and the resulting
DataFrame -- the only major thing still left on the table
Github user leifwalsh commented on the issue:
https://github.com/apache/spark/pull/15821
The next iteration of this for perf would likely involve generating the
arrow batches on executors and having the driver use the new streaming arrow
format to just forward this to python. In our e
Github user holdenk commented on the issue:
https://github.com/apache/spark/pull/15821
On a personal note, those benchmarks certainly look very exciting (<3 max
of with arrow less than min of without arrow) :)
It certainly seems it would probably be worth the review bandwidth
Github user BryanCutler commented on the issue:
https://github.com/apache/spark/pull/15821
Here are some rough benchmarks done locally on machine with 16GB mem and 8
cores, using Spark config defaults and taken from 50 trials of calling
`toPandas()` with and without Arrow enabled:
Github user BryanCutler commented on the issue:
https://github.com/apache/spark/pull/15821
This has been updated after integrating changes made with @icexelloss and
@wesm. There has been good progress made and it would be great if others could
take a look and review/test this out.
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/15821
Merged build finished. Test FAILed.
---
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
e
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/15821
Test FAILed.
Refer to this link for build results (access rights to CI server needed):
https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/71950/
Test FAILed.
---
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/15821
**[Test build #71950 has
finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/71950/testReport)**
for PR 15821 at commit
[`9bb75de`](https://github.com/apache/spark/commit/9
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/15821
**[Test build #71950 has
started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/71950/testReport)**
for PR 15821 at commit
[`9bb75de`](https://github.com/apache/spark/commit/9b
Github user BryanCutler commented on the issue:
https://github.com/apache/spark/pull/15821
That sounds good, I'm out of town atm but will update this tomorrow to get
some some more eyes on it.
On Jan 23, 2017 11:40 AM, "Li Jin" wrote:
> Bryan,
>
> I am wor
Github user icexelloss commented on the issue:
https://github.com/apache/spark/pull/15821
Bryan,
I am working on:
(1) Add more numbers to benchmark.py
(2) Add support for date/timestamp/binary type
(3) Fix memory leaking in the code.
All these should be don
Github user BryanCutler commented on the issue:
https://github.com/apache/spark/pull/15821
>Shall we update this PR to the latest and solicit from involvement from
Spark committers?
Yeah, I think it's about ready for that. After we integrate the latest
changes, I'll go over
Github user wesm commented on the issue:
https://github.com/apache/spark/pull/15821
Shall we update this PR to the latest and solicit from involvement from
Spark committers?
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as wel
Github user wesm commented on the issue:
https://github.com/apache/spark/pull/15821
OK, let's open pull requests into that branch to help with not stepping on
each other's toes. thank you
---
If your project is set up for it, you can reply to this email and have your
reply appear on
Github user BryanCutler commented on the issue:
https://github.com/apache/spark/pull/15821
@icexelloss, @wesm I branched off here for us to integrate our changes
https://github.com/BryanCutler/spark/tree/arrow-integration
cc @yinxusen
---
If your project is set up for it, you c
Github user icexelloss commented on the issue:
https://github.com/apache/spark/pull/15821
@BryanCutler , I have been working based on your branch here:
https://github.com/BryanCutler/spark/tree/wip-toPandas_with_arrow-SPARK-13534
Is this the right one?
---
If your projec
Github user BryanCutler commented on the issue:
https://github.com/apache/spark/pull/15821
Hi @wesm and @icexelloss , that sounds good on our end. @yinxusen has been
working on validating some basic conversion so far, but everything is still
very preliminary so it would be great to w
Github user wesm commented on the issue:
https://github.com/apache/spark/pull/15821
Related to this we'll also want to be able to precisely instrument and
benchmark the Dataset <-> Arrow conversion -- @icexelloss suggested might be
able to push down the conversion into the executors i
Github user wesm commented on the issue:
https://github.com/apache/spark/pull/15821
@BryanCutler I'm working with @icexelloss on my end to get involved in
this, we were going to start working on unit tests to validate converting each
of the Spark SQL data types to Arrow format while t
Github user wesm commented on the issue:
https://github.com/apache/spark/pull/15821
Luckily we are on the home stretch for making the Java and C++ libraries
binary compatible -- e.g. I'm working on automated testing today:
https://github.com/apache/arrow/pull/219
---
If your project
Github user BryanCutler commented on the issue:
https://github.com/apache/spark/pull/15821
Thanks @mariusvniekerk, as @holdenk said we are going to try to get
something basic working first and after we show some performance improvement,
we can follow up with more things
---
If your
Github user holdenk commented on the issue:
https://github.com/apache/spark/pull/15821
@mariusvniekerk I think just getting this working for local connection is
going to be hard so breaking up using arrow on the driver side into a separate
follow up piece of work would make sense.
-
Github user mariusvniekerk commented on the issue:
https://github.com/apache/spark/pull/15821
So this is very cool stuff.
Would it be reasonable to add some api pieces so that on the python side
things like DataFrame.mapPartitions makes use of Apache Arrow to lower the
seri
Github user BryanCutler commented on the issue:
https://github.com/apache/spark/pull/15821
Hey @holdenk, I just had this in to do my own testing and hadn't thought
about keeping the option, but if we do keep it then yeah you're right, it
would be better to default to the origina
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/15821
**[Test build #68954 has
finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/68954/consoleFull)**
for PR 15821 at commit
[`9191b96`](https://github.com/apache/spark/commit/
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/15821
Merged build finished. Test FAILed.
---
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
e
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/15821
Test FAILed.
Refer to this link for build results (access rights to CI server needed):
https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/68954/
Test FAILed.
---
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/15821
**[Test build #68954 has
started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/68954/consoleFull)**
for PR 15821 at commit
[`9191b96`](https://github.com/apache/spark/commit/9
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/15821
**[Test build #68812 has
finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/68812/consoleFull)**
for PR 15821 at commit
[`9191b96`](https://github.com/apache/spark/commit/
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/15821
Test FAILed.
Refer to this link for build results (access rights to CI server needed):
https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/68812/
Test FAILed.
---
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/15821
Merged build finished. Test FAILed.
---
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
e
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/15821
**[Test build #68812 has
started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/68812/consoleFull)**
for PR 15821 at commit
[`9191b96`](https://github.com/apache/spark/commit/9
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/15821
Merged build finished. Test FAILed.
---
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
e
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/15821
**[Test build #68806 has
finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/68806/consoleFull)**
for PR 15821 at commit
[`053e3a6`](https://github.com/apache/spark/commit/
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/15821
Test FAILed.
Refer to this link for build results (access rights to CI server needed):
https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/68806/
Test FAILed.
---
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/15821
**[Test build #68806 has
started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/68806/consoleFull)**
for PR 15821 at commit
[`053e3a6`](https://github.com/apache/spark/commit/0
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/15821
**[Test build #68427 has
finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/68427/consoleFull)**
for PR 15821 at commit
[`b06e11f`](https://github.com/apache/spark/commit/
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/15821
Build finished. Test FAILed.
---
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
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/15821
Test FAILed.
Refer to this link for build results (access rights to CI server needed):
https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/68427/
Test FAILed.
---
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/15821
**[Test build #68427 has
started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/68427/consoleFull)**
for PR 15821 at commit
[`b06e11f`](https://github.com/apache/spark/commit/b
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/15821
Merged build finished. Test FAILed.
---
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
e
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/15821
Test FAILed.
Refer to this link for build results (access rights to CI server needed):
https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/68425/
Test FAILed.
---
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/15821
**[Test build #68425 has
finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/68425/consoleFull)**
for PR 15821 at commit
[`3f855ec`](https://github.com/apache/spark/commit/
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/15821
**[Test build #68425 has
started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/68425/consoleFull)**
for PR 15821 at commit
[`3f855ec`](https://github.com/apache/spark/commit/3
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/15821
Test FAILed.
Refer to this link for build results (access rights to CI server needed):
https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/68381/
Test FAILed.
---
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/15821
**[Test build #68381 has
finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/68381/consoleFull)**
for PR 15821 at commit
[`4227ec6`](https://github.com/apache/spark/commit/
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/15821
Merged build finished. Test FAILed.
---
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
e
Github user BryanCutler commented on the issue:
https://github.com/apache/spark/pull/15821
the test currently fails with the stack trace
```
Traceback (most recent call last):
File "/home/bryan/git/spark/python/pyspark/sql/tests.py", line 2000, in
test_arrow_round_trip
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/15821
**[Test build #68381 has
started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/68381/consoleFull)**
for PR 15821 at commit
[`4227ec6`](https://github.com/apache/spark/commit/4
93 matches
Mail list logo
