Github user jkbradley commented on the pull request:
https://github.com/apache/spark/pull/8757#issuecomment-146733572
@brkyvz Can you please add "[MLLIB]" to the PR title?
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. I
Github user SparkQA commented on the pull request:
https://github.com/apache/spark/pull/8757#issuecomment-146724888
[Test build #1864 has
finished](https://amplab.cs.berkeley.edu/jenkins/job/NewSparkPullRequestBuilder/1864/console)
for PR 8757 at commit
[`ae98edc`](https://github
Github user jkbradley commented on the pull request:
https://github.com/apache/spark/pull/8757#issuecomment-146723162
Minor comments only. Other than that, it looks fine to me.
@avulanov In your "Block-diagonal matrix multiplication" tests, do you know
if data were shuffled d
Github user jkbradley commented on a diff in the pull request:
https://github.com/apache/spark/pull/8757#discussion_r41586262
--- Diff:
mllib/src/main/scala/org/apache/spark/mllib/linalg/distributed/BlockMatrix.scala
---
@@ -352,6 +353,30 @@ class BlockMatrix @Since("1.3.0") (
Github user jkbradley commented on a diff in the pull request:
https://github.com/apache/spark/pull/8757#discussion_r41586254
--- Diff:
mllib/src/main/scala/org/apache/spark/mllib/linalg/distributed/BlockMatrix.scala
---
@@ -352,6 +353,30 @@ class BlockMatrix @Since("1.3.0") (
Github user jkbradley commented on a diff in the pull request:
https://github.com/apache/spark/pull/8757#discussion_r41586259
--- Diff:
mllib/src/main/scala/org/apache/spark/mllib/linalg/distributed/BlockMatrix.scala
---
@@ -352,6 +353,30 @@ class BlockMatrix @Since("1.3.0") (
Github user jkbradley commented on a diff in the pull request:
https://github.com/apache/spark/pull/8757#discussion_r41586252
--- Diff:
mllib/src/main/scala/org/apache/spark/mllib/linalg/distributed/BlockMatrix.scala
---
@@ -60,6 +60,7 @@ private[mllib] class GridPartitioner(
Github user jkbradley commented on the pull request:
https://github.com/apache/spark/pull/8757#issuecomment-146720466
Making a pass now
---
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 fe
Github user SparkQA commented on the pull request:
https://github.com/apache/spark/pull/8757#issuecomment-146717274
[Test build #1864 has
started](https://amplab.cs.berkeley.edu/jenkins/job/NewSparkPullRequestBuilder/1864/consoleFull)
for PR 8757 at commit
[`ae98edc`](https://git
Github user avulanov commented on the pull request:
https://github.com/apache/spark/pull/8757#issuecomment-140554085
Thank you for the update. Indeed, the tests take finite time to finish now.
Let's add @mengxr to the discussion.
Distributed matrix multiplication makes sense w
Github user brkyvz commented on the pull request:
https://github.com/apache/spark/pull/8757#issuecomment-140470928
More results. Tests performed on 4 executors each with 30 GB RAM, and 4
cores each:
- Size of A: 1e5 x 1e5
- Size of B: 1e5 x 1e5
- Block Size: 1024
Github user brkyvz commented on the pull request:
https://github.com/apache/spark/pull/8757#issuecomment-140235259
@avulanov Feel free to benchmark it in anyway. The same code is also
useful. I'm interested in how it would scale, and how it would perform if the
matrix is fully dense.
Github user avulanov commented on the pull request:
https://github.com/apache/spark/pull/8757#issuecomment-140234877
@brkyvz Thank you for notifying me. I would be interested to benchmark this
PR. Should I use the same code from the mailing list? It can be found here as
well https://g
Github user AmplabJenkins commented on the pull request:
https://github.com/apache/spark/pull/8757#issuecomment-140218848
Test FAILed.
Refer to this link for build results (access rights to CI server needed):
https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/42
Github user AmplabJenkins commented on the pull request:
https://github.com/apache/spark/pull/8757#issuecomment-140218846
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
Github user AmplabJenkins commented on the pull request:
https://github.com/apache/spark/pull/8757#issuecomment-140215646
Merged build triggered.
---
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 h
Github user AmplabJenkins commented on the pull request:
https://github.com/apache/spark/pull/8757#issuecomment-140215666
Merged build started.
---
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
Github user SparkQA commented on the pull request:
https://github.com/apache/spark/pull/8757#issuecomment-140214626
[Test build #42448 has
finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/42448/console)
for PR 8757 at commit
[`8dac58f`](https://github.
Github user AmplabJenkins commented on the pull request:
https://github.com/apache/spark/pull/8757#issuecomment-140214630
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
Github user AmplabJenkins commented on the pull request:
https://github.com/apache/spark/pull/8757#issuecomment-140214633
Test FAILed.
Refer to this link for build results (access rights to CI server needed):
https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/42
Github user SparkQA commented on the pull request:
https://github.com/apache/spark/pull/8757#issuecomment-140214200
[Test build #42448 has
started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/42448/consoleFull)
for PR 8757 at commit
[`8dac58f`](https://gith
Github user AmplabJenkins commented on the pull request:
https://github.com/apache/spark/pull/8757#issuecomment-140212664
Merged build started.
---
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
Github user AmplabJenkins commented on the pull request:
https://github.com/apache/spark/pull/8757#issuecomment-140212640
Merged build triggered.
---
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 h
GitHub user brkyvz opened a pull request:
https://github.com/apache/spark/pull/8757
[SPARK-10599] Lower communication for block matrix multiplication
This PR aims to decrease communication costs in BlockMatrix multiplication
in two ways:
- Simulate the multiplication on the dri
24 matches
Mail list logo