[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-12 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/13704 **[Test build #62152 has started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/62152/consoleFull)** for PR 13704 at commit

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-11 Thread kiszk
Github user kiszk commented on the issue: https://github.com/apache/spark/pull/13704 @cloud-fan I added test cases. Could you review it again? --- 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] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-11 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/13704 Merged build finished. Test PASSed. --- 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

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-11 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/13704 Test PASSed. Refer to this link for build results (access rights to CI server needed): https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/62102/ Test PASSed. ---

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-11 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/13704 **[Test build #62102 has finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/62102/consoleFull)** for PR 13704 at commit

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-11 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/13704 **[Test build #62102 has started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/62102/consoleFull)** for PR 13704 at commit

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-11 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/13704 Test FAILed. Refer to this link for build results (access rights to CI server needed): https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/62097/ Test FAILed. ---

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-11 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/13704 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

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-11 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/13704 **[Test build #62097 has finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/62097/consoleFull)** for PR 13704 at commit

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-11 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/13704 **[Test build #62097 has started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/62097/consoleFull)** for PR 13704 at commit

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-11 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/13704 Merged build finished. Test PASSed. --- 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

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-11 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/13704 Test PASSed. Refer to this link for build results (access rights to CI server needed): https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/62087/ Test PASSed. ---

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-11 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/13704 **[Test build #62087 has finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/62087/consoleFull)** for PR 13704 at commit

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-11 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/13704 Test PASSed. Refer to this link for build results (access rights to CI server needed): https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/62084/ Test PASSed. ---

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-11 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/13704 Merged build finished. Test PASSed. --- 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

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-11 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/13704 **[Test build #62084 has finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/62084/consoleFull)** for PR 13704 at commit

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-11 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/13704 **[Test build #62084 has started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/62084/consoleFull)** for PR 13704 at commit

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-11 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/13704 **[Test build #62087 has started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/62087/consoleFull)** for PR 13704 at commit

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-10 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/13704 Test PASSed. Refer to this link for build results (access rights to CI server needed): https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/62071/ Test PASSed. ---

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-10 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/13704 Merged build finished. Test PASSed. --- 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

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-10 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/13704 **[Test build #62071 has finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/62071/consoleFull)** for PR 13704 at commit

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-10 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/13704 **[Test build #62071 has started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/62071/consoleFull)** for PR 13704 at commit

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-10 Thread kiszk
Github user kiszk commented on the issue: https://github.com/apache/spark/pull/13704 @cloud-fan could you please review this? As you pointed, I also changed code related to `cast`. I added benchmark results, too. --- If your project is set up for it, you can reply to this email and

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-10 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/13704 Test PASSed. Refer to this link for build results (access rights to CI server needed): https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/62057/ Test PASSed. ---

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-10 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/13704 Merged build finished. Test PASSed. --- 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

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-10 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/13704 **[Test build #62057 has finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/62057/consoleFull)** for PR 13704 at commit

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-10 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/13704 **[Test build #62057 has started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/62057/consoleFull)** for PR 13704 at commit

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-10 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/13704 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

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-10 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/13704 Test FAILed. Refer to this link for build results (access rights to CI server needed): https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/62050/ Test FAILed. ---

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-10 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/13704 **[Test build #62050 has finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/62050/consoleFull)** for PR 13704 at commit

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-10 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/13704 **[Test build #62050 has started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/62050/consoleFull)** for PR 13704 at commit

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-03 Thread kiszk
Github user kiszk commented on the issue: https://github.com/apache/spark/pull/13704 @cloud-fan , regarding a ```cast```, I think that you are correct. When we insert a ```cast```, I pass information on ```ArrayType.containsNull``` to ```Cast``` now. In this case, it is ```false```.

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-03 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/13704 Test PASSed. Refer to this link for build results (access rights to CI server needed): https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/61677/ Test PASSed. ---

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-03 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/13704 Merged build finished. Test PASSed. --- 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

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-03 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/13704 **[Test build #61677 has finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/61677/consoleFull)** for PR 13704 at commit

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-02 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/13704 **[Test build #61677 has started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/61677/consoleFull)** for PR 13704 at commit

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-01 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/13704 Test FAILed. Refer to this link for build results (access rights to CI server needed): https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/61630/ Test FAILed. ---

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-01 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/13704 **[Test build #61630 has finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/61630/consoleFull)** for PR 13704 at commit

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-01 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/13704 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

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-01 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/13704 Test FAILed. Refer to this link for build results (access rights to CI server needed): https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/61629/ Test FAILed. ---

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-01 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/13704 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

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-01 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/13704 **[Test build #61629 has finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/61629/consoleFull)** for PR 13704 at commit

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-01 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/13704 **[Test build #61630 has started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/61630/consoleFull)** for PR 13704 at commit

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-01 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/13704 **[Test build #61629 has started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/61629/consoleFull)** for PR 13704 at commit

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-07-01 Thread kiszk
Github user kiszk commented on the issue: https://github.com/apache/spark/pull/13704 @cloud-fan, I checked the following >Let me check which code portion inserts Cast in this tree. IIUC, [this

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-06-28 Thread kiszk
Github user kiszk commented on the issue: https://github.com/apache/spark/pull/13704 Let me check which code portion inserts ```Cast``` in this tree. > And we do have a chance to optimize it: if the target array type's element type is primitive and the input array type's

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-06-28 Thread cloud-fan
Github user cloud-fan commented on the issue: https://github.com/apache/spark/pull/13704 > the `Cast` performs conversion from `UnsafeArrayData` to `GenericArrayData`. `Cast` is used to cast one type to another, not cast one kind of data representation to another, so I don't

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-06-28 Thread kiszk
Github user kiszk commented on the issue: https://github.com/apache/spark/pull/13704 Regarding the plan tree printout (I removed my debug information), the ```Cast``` performs conversion from ```UnsafeArrayData``` to ```GenericArrayData``` since the target type of ```Cast``` is

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-06-28 Thread cloud-fan
Github user cloud-fan commented on the issue: https://github.com/apache/spark/pull/13704 From the plan tree given by you, the `Cast` is casting `value#63` to array type, not `ObjectType(classOf[Array[Double]])`, there should be some other reason that we need to find out(maybe it's

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-06-28 Thread kiszk
Github user kiszk commented on the issue: https://github.com/apache/spark/pull/13704 I agree that ```value#63``` is double array type. ```value#63``` is stored as ```UnsafeArrayData```. `` in ```MapElements , obj#67: double``` is represented by Java byte code instead of

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-06-28 Thread cloud-fan
Github user cloud-fan commented on the issue: https://github.com/apache/spark/pull/13704 Sorry I should say it more explicitly: `value#63` is already double array type. `UnsafeArrayData` is an internal data representation of array type, so it seems weird to have the `Cast` there.

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-06-27 Thread kiszk
Github user kiszk commented on the issue: https://github.com/apache/spark/pull/13704 @cloud-fan I think ```value#63``` is ```UnsafeArrayData```. When I ran a DataFrame program, I got the following trees. Since operations for DataFrame access data in UnsafeArrayData, I think that

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-06-27 Thread cloud-fan
Github user cloud-fan commented on the issue: https://github.com/apache/spark/pull/13704 I'm wondering why we have this `Cast` in `DeserializeToObject`. Isn't the `value#63` already a double array? --- If your project is set up for it, you can reply to this email and have your reply

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-06-27 Thread kiszk
Github user kiszk commented on the issue: https://github.com/apache/spark/pull/13704 @cloud-fan I updated the PR description by adding plan trees. --- 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

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-06-27 Thread cloud-fan
Github user cloud-fan commented on the issue: https://github.com/apache/spark/pull/13704 can you also put the plan tree of the example program in PR description? --- 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

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-06-16 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/13704 Test PASSed. Refer to this link for build results (access rights to CI server needed): https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/60636/ Test PASSed. ---

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-06-16 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/13704 Merged build finished. Test PASSed. --- 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

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-06-16 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/13704 **[Test build #60636 has finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/60636/consoleFull)** for PR 13704 at commit

[GitHub] spark issue #13704: [SPARK-15985][SQL] Reduce runtime overhead of a program ...

2016-06-16 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/13704 **[Test build #60636 has started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/60636/consoleFull)** for PR 13704 at commit