Github user srowen commented on the issue:
https://github.com/apache/spark/pull/19232
I'm gonna merge this as a non-trivial win. I think the benefit outweighs
concerns. We can add to the change later.
---
-
To unsub
Github user sethah commented on the issue:
https://github.com/apache/spark/pull/19232
Sure, we all agree there is a mechanism for avoiding overhead. However,
performance tests are very tricky things, 5% is not a huge improvement, and
hard-coding the aggregation depth to `2` limits the
Github user srowen commented on the issue:
https://github.com/apache/spark/pull/19232
Yeah I wonder if this slows things down for smaller data sets, because of
the extra levels and checks, but then again, when the aggregation is small,
anything's similarly fast. The default depth is s
Github user sethah commented on the issue:
https://github.com/apache/spark/pull/19232
I'm not really aware of situations where it would be detrimental, since it
has a mechanism for avoiding the intermediate stages when it doesn't make
sense. However, one of the big advantages of `tree
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/19232
Test PASSed.
Refer to this link for build results (access rights to CI server needed):
https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/81779/
Test PASSed.
---
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/19232
Merged build finished. Test PASSed.
---
-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional comma
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/19232
**[Test build #81779 has
finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/81779/testReport)**
for PR 19232 at commit
[`0c3ed2d`](https://github.com/apache/spark/commit/0
Github user srowen commented on the issue:
https://github.com/apache/spark/pull/19232
@sethah I feel like we talked about something like this in BigDL. Is there
really any downside to treeAggregate?
---
-
To unsubsc
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/19232
**[Test build #81779 has
started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/81779/testReport)**
for PR 19232 at commit
[`0c3ed2d`](https://github.com/apache/spark/commit/0c
Github user zhengruifeng commented on the issue:
https://github.com/apache/spark/pull/19232
ping @yanboliang
---
-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: review
10 matches
Mail list logo