## What changes were proposed in this pull request?
Implement an image schema datasource.
This image datasource support:
- partition discovery (loading partitioned images)
- dropImageFailures (the same behavior with `ImageSchema.readImage`)
- path wildcard matching (the same behavior with `ImageSchema.readImage`)
- loading recursively from directory (different from `ImageSchema.readImage`, but use such path: `/path/to/dir/**`)
This datasource **NOT** support:
- specify `numPartitions` (it will be determined by datasource automatically)
- sampling (you can use `df.sample` later but the sampling operator won't be pushdown to datasource)
## How was this patch tested?
Unit tests.
## Benchmark
I benchmark and compare the cost time between old `ImageSchema.read` API and my image datasource.
**cluster**: 4 nodes, each with 64GB memory, 8 cores CPU
**test dataset**: Flickr8k_Dataset (about 8091 images)
**time cost**:
- My image datasource time (automatically generate 258 partitions): 38.04s
- `ImageSchema.read` time (set 16 partitions): 68.4s
- `ImageSchema.read` time (set 258 partitions): 90.6s
**time cost when increase image number by double (clone Flickr8k_Dataset and loads double number images)**:
- My image datasource time (automatically generate 515 partitions): 95.4s
- `ImageSchema.read` (set 32 partitions): 109s
- `ImageSchema.read` (set 515 partitions): 105s
So we can see that my image datasource implementation (this PR) bring some performance improvement compared against old`ImageSchema.read` API.
Closes#22328 from WeichenXu123/image_datasource.
Authored-by: WeichenXu <weichen.xu@databricks.com>
Signed-off-by: Xiangrui Meng <meng@databricks.com>
## What changes were proposed in this pull request?
This is a follow-up of #22313 and aim to ignore the micro benchmark test which takes over 2 minutes in Jenkins.
- https://amplab.cs.berkeley.edu/jenkins/view/Spark%20QA%20Test%20(Dashboard)/job/spark-master-test-sbt-hadoop-2.6/4939/consoleFull
## How was this patch tested?
The test case should be ignored in Jenkins.
```
[info] FilterPushdownBenchmark:
...
[info] - Pushdown benchmark with many filters !!! IGNORED !!!
```
Closes#22336 from dongjoon-hyun/SPARK-25306-2.
Authored-by: Dongjoon Hyun <dongjoon@apache.org>
Signed-off-by: Xiao Li <gatorsmile@gmail.com>
The problem occurs because stage object is removed from liveStages in
AppStatusListener onStageCompletion. Because of this any onTaskEnd event
received after onStageCompletion event do not update stage metrics.
The fix is to retain stage objects in liveStages until all tasks are complete.
1. Fixed the reproducible example posted in the JIRA
2. Added unit test
Closes#22209 from ankuriitg/ankurgupta/SPARK-24415.
Authored-by: ankurgupta <ankur.gupta@cloudera.com>
Signed-off-by: Marcelo Vanzin <vanzin@cloudera.com>
## What changes were proposed in this pull request?
Add a new metric to measure the executor's process (JVM) CPU time.
## How was this patch tested?
Manually tested on a Spark cluster (see SPARK-25228 for an example screenshot).
Closes#22218 from LucaCanali/AddExecutrCPUTimeMetric.
Authored-by: LucaCanali <luca.canali@cern.ch>
Signed-off-by: Sean Owen <sean.owen@databricks.com>
## What changes were proposed in this pull request?
This is a followup of https://github.com/apache/spark/pull/22259 .
Scala case class has a wide surface: apply, unapply, accessors, copy, etc.
In https://github.com/apache/spark/pull/22259 , we change the type of `UserDefinedFunction.inputTypes` from `Option[Seq[DataType]]` to `Option[Seq[Schema]]`. This breaks backward compatibility.
This PR changes the type back, and use a `var` to keep the new nullable info.
## How was this patch tested?
N/A
Closes#22319 from cloud-fan/revert.
Authored-by: Wenchen Fan <wenchen@databricks.com>
Signed-off-by: Wenchen Fan <wenchen@databricks.com>
## What changes were proposed in this pull request?
Revert SPARK-24863 (#21819) and SPARK-24748 (#21721) as per discussion in #21721. We will revisit them when the data source v2 APIs are out.
## How was this patch tested?
Jenkins
Closes#22334 from zsxwing/revert-SPARK-24863-SPARK-24748.
Authored-by: Shixiong Zhu <zsxwing@gmail.com>
Signed-off-by: Wenchen Fan <wenchen@databricks.com>
## What changes were proposed in this pull request?
The configuration parameter "spark.shuffle.service.enabled" has defined in `package.scala`, and it is also used in many place, so we can replace it with `SHUFFLE_SERVICE_ENABLED`.
and unified this configuration parameter "spark.shuffle.service.port" together.
## How was this patch tested?
N/A
Closes#22306 from 10110346/unifiedserviceenable.
Authored-by: liuxian <liu.xian3@zte.com.cn>
Signed-off-by: Wenchen Fan <wenchen@databricks.com>
## What changes were proposed in this pull request?
In both ORC data sources, `createFilter` function has exponential time complexity due to its skewed filter tree generation. This PR aims to improve it by using new `buildTree` function.
**REPRODUCE**
```scala
// Create and read 1 row table with 1000 columns
sql("set spark.sql.orc.filterPushdown=true")
val selectExpr = (1 to 1000).map(i => s"id c$i")
spark.range(1).selectExpr(selectExpr: _*).write.mode("overwrite").orc("/tmp/orc")
print(s"With 0 filters, ")
spark.time(spark.read.orc("/tmp/orc").count)
// Increase the number of filters
(20 to 30).foreach { width =>
val whereExpr = (1 to width).map(i => s"c$i is not null").mkString(" and ")
print(s"With $width filters, ")
spark.time(spark.read.orc("/tmp/orc").where(whereExpr).count)
}
```
**RESULT**
```scala
With 0 filters, Time taken: 653 ms
With 20 filters, Time taken: 962 ms
With 21 filters, Time taken: 1282 ms
With 22 filters, Time taken: 1982 ms
With 23 filters, Time taken: 3855 ms
With 24 filters, Time taken: 6719 ms
With 25 filters, Time taken: 12669 ms
With 26 filters, Time taken: 25032 ms
With 27 filters, Time taken: 49585 ms
With 28 filters, Time taken: 98980 ms // over 1 min 38 seconds
With 29 filters, Time taken: 198368 ms // over 3 mins
With 30 filters, Time taken: 393744 ms // over 6 mins
```
**AFTER THIS PR**
```scala
With 0 filters, Time taken: 774 ms
With 20 filters, Time taken: 601 ms
With 21 filters, Time taken: 399 ms
With 22 filters, Time taken: 679 ms
With 23 filters, Time taken: 363 ms
With 24 filters, Time taken: 342 ms
With 25 filters, Time taken: 336 ms
With 26 filters, Time taken: 352 ms
With 27 filters, Time taken: 322 ms
With 28 filters, Time taken: 302 ms
With 29 filters, Time taken: 307 ms
With 30 filters, Time taken: 301 ms
```
## How was this patch tested?
Pass the Jenkins with newly added test cases.
Closes#22313 from dongjoon-hyun/SPARK-25306.
Authored-by: Dongjoon Hyun <dongjoon@apache.org>
Signed-off-by: Wenchen Fan <wenchen@databricks.com>
## What changes were proposed in this pull request?
I made one pass over barrier APIs added to Spark 2.4 and updates some scopes and docs. I will update Python docs once Scala doc was reviewed.
One major issue is that `BarrierTaskContext` implements `TaskContextImpl` that exposes some public methods. And internally there were several direct references to `TaskContextImpl` methods instead of `TaskContext`. This PR moved some methods from `TaskContextImpl` to `TaskContext`, remaining package private, and used delegate methods to avoid inheriting `TaskContextImp` and exposing unnecessary APIs.
TODOs:
- [x] scala doc
- [x] python doc (#22261 ).
Closes#22240 from mengxr/SPARK-25248.
Authored-by: Xiangrui Meng <meng@databricks.com>
Signed-off-by: Xiangrui Meng <meng@databricks.com>
## What changes were proposed in this pull request?
Previously in `TakeOrderedAndProjectSuite` the SparkSession will not get recycled when the test suite finishes.
## How was this patch tested?
N/A
Closes#22330 from jiangxb1987/SPARK-19355.
Authored-by: Xingbo Jiang <xingbo.jiang@databricks.com>
Signed-off-by: Xiao Li <gatorsmile@gmail.com>
## What changes were proposed in this pull request?
This PR integrates handling of `UnsafeArrayData` and `GenericArrayData` into one. The current `CodeGenerator.createUnsafeArray` handles only allocation of `UnsafeArrayData`.
This PR introduces a new method `createArrayData` that returns a code to allocate `UnsafeArrayData` or `GenericArrayData` and to assign a value into the allocated array.
This PR also reduce the size of generated code by calling a runtime helper.
This PR replaced `createArrayData` with `createUnsafeArray`. This PR also refactor `ArraySetLike` that can be used for `ArrayDistinct`, too.
This PR also refactors`ArrayDistinct` to use `ArraryBuilder`.
## How was this patch tested?
Existing tests
Closes#21912 from kiszk/SPARK-24962.
Lead-authored-by: Kazuaki Ishizaki <ishizaki@jp.ibm.com>
Co-authored-by: Takuya UESHIN <ueshin@happy-camper.st>
Signed-off-by: Wenchen Fan <wenchen@databricks.com>
## What changes were proposed in this pull request?
This PR fixes a problem that `ArraysOverlap` function throws a `CompilationException` with non-nullable array type.
The following is the stack trace of the original problem:
```
Code generation of arrays_overlap([1,2,3], [4,5,3]) failed:
java.util.concurrent.ExecutionException: org.codehaus.commons.compiler.CompileException: File 'generated.java', Line 56, Column 11: failed to compile: org.codehaus.commons.compiler.CompileException: File 'generated.java', Line 56, Column 11: Expression "isNull_0" is not an rvalue
java.util.concurrent.ExecutionException: org.codehaus.commons.compiler.CompileException: File 'generated.java', Line 56, Column 11: failed to compile: org.codehaus.commons.compiler.CompileException: File 'generated.java', Line 56, Column 11: Expression "isNull_0" is not an rvalue
at com.google.common.util.concurrent.AbstractFuture$Sync.getValue(AbstractFuture.java:306)
at com.google.common.util.concurrent.AbstractFuture$Sync.get(AbstractFuture.java:293)
at com.google.common.util.concurrent.AbstractFuture.get(AbstractFuture.java:116)
at com.google.common.util.concurrent.Uninterruptibles.getUninterruptibly(Uninterruptibles.java:135)
at com.google.common.cache.LocalCache$Segment.getAndRecordStats(LocalCache.java:2410)
at com.google.common.cache.LocalCache$Segment.loadSync(LocalCache.java:2380)
at com.google.common.cache.LocalCache$Segment.lockedGetOrLoad(LocalCache.java:2342)
at com.google.common.cache.LocalCache$Segment.get(LocalCache.java:2257)
at com.google.common.cache.LocalCache.get(LocalCache.java:4000)
at com.google.common.cache.LocalCache.getOrLoad(LocalCache.java:4004)
at com.google.common.cache.LocalCache$LocalLoadingCache.get(LocalCache.java:4874)
at org.apache.spark.sql.catalyst.expressions.codegen.CodeGenerator$.compile(CodeGenerator.scala:1305)
at org.apache.spark.sql.catalyst.expressions.codegen.GenerateMutableProjection$.create(GenerateMutableProjection.scala:143)
at org.apache.spark.sql.catalyst.expressions.codegen.GenerateMutableProjection$.create(GenerateMutableProjection.scala:48)
at org.apache.spark.sql.catalyst.expressions.codegen.GenerateMutableProjection$.create(GenerateMutableProjection.scala:32)
at org.apache.spark.sql.catalyst.expressions.codegen.CodeGenerator.generate(CodeGenerator.scala:1260)
```
## How was this patch tested?
Added test in `CollectionExpressionSuite`.
Closes#22317 from kiszk/SPARK-25310.
Authored-by: Kazuaki Ishizaki <ishizaki@jp.ibm.com>
Signed-off-by: Takuya UESHIN <ueshin@databricks.com>
## What changes were proposed in this pull request?
Invoking ArrayContains function with non nullable array type throws the following error in the code generation phase. Below is the error snippet.
```SQL
Code generation of array_contains([1,2,3], 1) failed:
java.util.concurrent.ExecutionException: org.codehaus.commons.compiler.CompileException: File 'generated.java', Line 40, Column 11: failed to compile: org.codehaus.commons.compiler.CompileException: File 'generated.java', Line 40, Column 11: Expression "isNull_0" is not an rvalue
java.util.concurrent.ExecutionException: org.codehaus.commons.compiler.CompileException: File 'generated.java', Line 40, Column 11: failed to compile: org.codehaus.commons.compiler.CompileException: File 'generated.java', Line 40, Column 11: Expression "isNull_0" is not an rvalue
at com.google.common.util.concurrent.AbstractFuture$Sync.getValue(AbstractFuture.java:306)
at com.google.common.util.concurrent.AbstractFuture$Sync.get(AbstractFuture.java:293)
at com.google.common.util.concurrent.AbstractFuture.get(AbstractFuture.java:116)
at com.google.common.util.concurrent.Uninterruptibles.getUninterruptibly(Uninterruptibles.java:135)
at com.google.common.cache.LocalCache$Segment.getAndRecordStats(LocalCache.java:2410)
at com.google.common.cache.LocalCache$Segment.loadSync(LocalCache.java:2380)
at com.google.common.cache.LocalCache$Segment.lockedGetOrLoad(LocalCache.java:2342)
at com.google.common.cache.LocalCache$Segment.get(LocalCache.java:2257)
at com.google.common.cache.LocalCache.get(LocalCache.java:4000)
at com.google.common.cache.LocalCache.getOrLoad(LocalCache.java:4004)
at com.google.common.cache.LocalCache$LocalLoadingCache.get(LocalCache.java:4874)
at org.apache.spark.sql.catalyst.expressions.codegen.CodeGenerator$.compile(CodeGenerator.scala:1305)
```
## How was this patch tested?
Added test in CollectionExpressionSuite.
Closes#22315 from dilipbiswal/SPARK-25308.
Authored-by: Dilip Biswal <dbiswal@us.ibm.com>
Signed-off-by: Takuya UESHIN <ueshin@databricks.com>
## What changes were proposed in this pull request?
Improve build for Scala 2.12. Current build for sbt fails on the subproject `repl`:
```
[info] Compiling 6 Scala sources to /Users/rendong/wdi/spark/repl/target/scala-2.12/classes...
[error] /Users/rendong/wdi/spark/repl/scala-2.11/src/main/scala/org/apache/spark/repl/SparkILoopInterpreter.scala:80: overriding lazy value importableSymbolsWithRenames in class ImportHandler of type List[(this.intp.global.Symbol, this.intp.global.Name)];
[error] lazy value importableSymbolsWithRenames needs `override' modifier
[error] lazy val importableSymbolsWithRenames: List[(Symbol, Name)] = {
[error] ^
[warn] /Users/rendong/wdi/spark/repl/src/main/scala/org/apache/spark/repl/SparkILoop.scala:53: variable addedClasspath in class ILoop is deprecated (since 2.11.0): use reset, replay or require to update class path
[warn] if (addedClasspath != "") {
[warn] ^
[warn] /Users/rendong/wdi/spark/repl/src/main/scala/org/apache/spark/repl/SparkILoop.scala:54: variable addedClasspath in class ILoop is deprecated (since 2.11.0): use reset, replay or require to update class path
[warn] settings.classpath append addedClasspath
[warn] ^
[warn] two warnings found
[error] one error found
[error] (repl/compile:compileIncremental) Compilation failed
[error] Total time: 93 s, completed 2018-9-3 10:07:26
```
## How was this patch tested?
```
./dev/change-scala-version.sh 2.12
## For Maven
./build/mvn -Pscala-2.12 [mvn commands]
## For SBT
sbt -Dscala.version=2.12.6
```
Closes#22310 from sadhen/SPARK-25298.
Authored-by: Darcy Shen <sadhen@zoho.com>
Signed-off-by: Sean Owen <sean.owen@databricks.com>
## What changes were proposed in this pull request?
The "date function on DataFrame" test fails consistently on my laptop. In this PR
i am fixing it by changing the way we compare the two timestamp values. With this change i am able to run the tests clean.
## How was this patch tested?
Fixed the failing test.
Author: Dilip Biswal <dbiswal@us.ibm.com>
Closes#22274 from dilipbiswal/r-sql-test-fix2.
## What changes were proposed in this pull request?
remove test-2.10.jar and add test-2.12.jar.
## How was this patch tested?
```
$ sbt -Dscala-2.12
> ++ 2.12.6
> project hive
> testOnly *HiveSparkSubmitSuite -- -z "8489"
```
Closes#22308 from sadhen/SPARK-8489-FOLLOWUP.
Authored-by: Darcy Shen <sadhen@zoho.com>
Signed-off-by: Sean Owen <sean.owen@databricks.com>
## What changes were proposed in this pull request?
Add the R version of array_intersect/array_except/array_union/shuffle
## How was this patch tested?
Add test in test_sparkSQL.R
Author: Huaxin Gao <huaxing@us.ibm.com>
Closes#22291 from huaxingao/spark-25007.
## What changes were proposed in this pull request?
The PR adds the lift measure to Association rules.
## How was this patch tested?
existing and modified UTs
Closes#22236 from mgaido91/SPARK-10697.
Authored-by: Marco Gaido <marcogaido91@gmail.com>
Signed-off-by: Sean Owen <sean.owen@databricks.com>
## What changes were proposed in this pull request?
Currently, when FDR is used for `ChiSqSelector` and no feature is selected an exception is thrown because the max operation fails.
The PR fixes the problem by handling this case and returning an empty array in that case, as sklearn (which was the reference for the initial implementation of FDR) does.
## How was this patch tested?
added UT
Closes#22303 from mgaido91/SPARK-25289.
Authored-by: Marco Gaido <marcogaido91@gmail.com>
Signed-off-by: Sean Owen <sean.owen@databricks.com>
## What changes were proposed in this pull request?
`BytesToBytesMapOnHeapSuite`.`randomizedStressTest` caused `OutOfMemoryError` on several test runs. Seems better to reduce memory usage in this test.
## How was this patch tested?
Unit tests.
Closes#22297 from viirya/SPARK-25290.
Authored-by: Liang-Chi Hsieh <viirya@gmail.com>
Signed-off-by: Wenchen Fan <wenchen@databricks.com>
## What changes were proposed in this pull request?
This PR works for one of TODOs in `GenerateUnsafeProjection` "if the nullability of field is correct, we can use it to save null check" to simplify generated code.
When `nullable=false` in `DataType`, `GenerateUnsafeProjection` removed code for null checks in the generated Java code.
## How was this patch tested?
Added new test cases into `GenerateUnsafeProjectionSuite`
Closes#20637 from kiszk/SPARK-23466.
Authored-by: Kazuaki Ishizaki <ishizaki@jp.ibm.com>
Signed-off-by: Takuya UESHIN <ueshin@databricks.com>
## What changes were proposed in this pull request?
Fixes the issue brought up in https://github.com/GoogleCloudPlatform/spark-on-k8s-operator/issues/273 where the arguments were being comma-delineated, which was incorrect wrt to the PythonRunner and RRunner.
## How was this patch tested?
Modified unit test to test this change.
Author: Ilan Filonenko <if56@cornell.edu>
Closes#22257 from ifilonenko/SPARK-25264.
## What changes were proposed in this pull request?
I propose to remove one of `parmap` methods which accepts an execution context as a parameter. The method should be removed to eliminate any deadlocks that can occur if `parmap` is called recursively on thread pools restricted by size.
Closes#22292 from MaxGekk/remove-overloaded-parmap.
Authored-by: Maxim Gekk <maxim.gekk@databricks.com>
Signed-off-by: Xiao Li <gatorsmile@gmail.com>
## What changes were proposed in this pull request?
Move the output verification of Explain test cases to a new suite ExplainSuite.
## How was this patch tested?
N/A
Closes#22300 from gatorsmile/test3200.
Authored-by: Xiao Li <gatorsmile@gmail.com>
Signed-off-by: Xiao Li <gatorsmile@gmail.com>
## What changes were proposed in this pull request?
As described in [SPARK-25261](https://issues.apache.org/jira/projects/SPARK/issues/SPARK-25261)，the unit of spark.executor.memory and spark.driver.memory is parsed as bytes in some cases if no unit specified, while in https://spark.apache.org/docs/latest/configuration.html#application-properties, they are descibed as MiB, which may lead to some misunderstandings.
## How was this patch tested?
N/A
Closes#22252 from ivoson/branch-correct-configuration.
Lead-authored-by: huangtengfei02 <huangtengfei02@baidu.com>
Co-authored-by: Huang Tengfei <tengfei.h@gmail.com>
Signed-off-by: Sean Owen <sean.owen@databricks.com>
## What changes were proposed in this pull request?
Currently, filter pushdown will not work if Parquet schema and Hive metastore schema are in different letter cases even spark.sql.caseSensitive is false.
Like the below case:
```scala
spark.sparkContext.hadoopConfiguration.setInt("parquet.block.size", 8 * 1024 * 1024)
spark.range(1, 40 * 1024 * 1024, 1, 1).sortWithinPartitions("id").write.parquet("/tmp/t")
sql("CREATE TABLE t (ID LONG) USING parquet LOCATION '/tmp/t'")
sql("select * from t where id < 100L").write.csv("/tmp/id")
```
Although filter "ID < 100L" is generated by Spark, it fails to pushdown into parquet actually, Spark still does the full table scan when reading.
This PR provides a case-insensitive field resolution to make it work.
Before - "ID < 100L" fail to pushedown:
<img width="273" alt="screen shot 2018-08-23 at 10 08 26 pm" src="https://user-images.githubusercontent.com/2989575/44530558-40ef8b00-a721-11e8-8abc-7f97671590d3.png">
After - "ID < 100L" pushedown sucessfully:
<img width="267" alt="screen shot 2018-08-23 at 10 08 40 pm" src="https://user-images.githubusercontent.com/2989575/44530567-44831200-a721-11e8-8634-e9f664b33d39.png">
## How was this patch tested?
Added UTs.
Closes#22197 from yucai/SPARK-25207.
Authored-by: yucai <yyu1@ebay.com>
Signed-off-by: Wenchen Fan <wenchen@databricks.com>
## What changes were proposed in this pull request?
Switch `org.apache.hive.service.server.HiveServer2` to register its shutdown callback with Spark's `ShutdownHookManager`, rather than direct with the Java Runtime callback.
This avoids race conditions in shutdown where the filesystem is shutdown before the flush/write/rename of the event log is completed, particularly on object stores where the write and rename can be slow.
## How was this patch tested?
There's no explicit unit for test this, which is consistent with every other shutdown hook in the codebase.
* There's an implicit test when the scalatest process is halted.
* More manual/integration testing is needed.
HADOOP-15679 has added the ability to explicitly execute the hadoop shutdown hook sequence which spark uses; that could be stabilized for testing if desired, after which all the spark hooks could be tested. Until then: external system tests only.
Author: Steve Loughran <stevel@hortonworks.com>
Closes#22186 from steveloughran/BUG/SPARK-25183-shutdown.
## What changes were proposed in this pull request?
### For `SPARK-5775 read array from partitioned_parquet_with_key_and_complextypes`:
scala2.12
```
scala> (1 to 10).toString
res4: String = Range 1 to 10
```
scala2.11
```
scala> (1 to 10).toString
res2: String = Range(1, 2, 3, 4, 5, 6, 7, 8, 9, 10)
```
And
```
def prepareAnswer(answer: Seq[Row], isSorted: Boolean): Seq[Row] = {
val converted: Seq[Row] = answer.map(prepareRow)
if (!isSorted) converted.sortBy(_.toString()) else converted
}
```
sortBy `_.toString` is not a good idea.
### Other failures are caused by
```
Array(Int.box(1)).toSeq == Array(Double.box(1.0)).toSeq
```
It is false in 2.12.2 + and is true in 2.11.x , 2.12.0, 2.12.1
## How was this patch tested?
This is a patch on a specific unit test.
Closes#22264 from sadhen/SPARK25256.
Authored-by: 忍冬 <rendong@wacai.com>
Signed-off-by: Sean Owen <sean.owen@databricks.com>
## What changes were proposed in this pull request?
Add an up-front check that `JIRA_USERNAME` and `JIRA_PASSWORD` have been set. If they haven't, ask user if they want to continue. This prevents the JIRA state update from failing at the very end of the process because user forgot to set these environment variables.
## How was this patch tested?
I ran the script with environment vars set, and unset, to verify it works as specified.
Please review http://spark.apache.org/contributing.html before opening a pull request.
Closes#22294 from erikerlandson/spark-25287.
Authored-by: Erik Erlandson <eerlands@redhat.com>
Signed-off-by: Erik Erlandson <eerlands@redhat.com>
## What changes were proposed in this pull request?
Add a PAM configuration in k8s dockerfile to require authentication into wheel to run as `su`
## How was this patch tested?
Verify against CI that PAM config succeeds & causes no regressions
Closes#22285 from erikerlandson/spark-25275.
Authored-by: Erik Erlandson <eerlands@redhat.com>
Signed-off-by: Erik Erlandson <eerlands@redhat.com>
## What changes were proposed in this pull request?
Error messages from https://amplab.cs.berkeley.edu/jenkins/view/Spark%20QA%20Test/job/spark-master-test-maven-hadoop-2.7-ubuntu-scala-2.12/183/
```
[INFO] --- scala-maven-plugin:3.2.2:compile (scala-compile-first) spark-repl_2.12 ---
[INFO] Using zinc server for incremental compilation
[warn] Pruning sources from previous analysis, due to incompatible CompileSetup.
[info] Compiling 6 Scala sources to /home/jenkins/workspace/spark-master-test-maven-hadoop-2.7-ubuntu-scala-2.12/repl/target/scala-2.12/classes...
[error] /home/jenkins/workspace/spark-master-test-maven-hadoop-2.7-ubuntu-scala-2.12/repl/scala-2.11/src/main/scala/org/apache/spark/repl/SparkILoopInterpreter.scala:80: overriding lazy value importableSymbolsWithRenames in class ImportHandler of type List[(this.intp.global.Symbol, this.intp.global.Name)];
[error] lazy value importableSymbolsWithRenames needs `override' modifier
[error] lazy val importableSymbolsWithRenames: List[(Symbol, Name)] = {
[error] ^
[warn] /home/jenkins/workspace/spark-master-test-maven-hadoop-2.7-ubuntu-scala-2.12/repl/src/main/scala/org/apache/spark/repl/SparkILoop.scala:53: variable addedClasspath in class ILoop is deprecated (since 2.11.0): use reset, replay or require to update class path
[warn] if (addedClasspath != "") {
[warn] ^
[warn] /home/jenkins/workspace/spark-master-test-maven-hadoop-2.7-ubuntu-scala-2.12/repl/src/main/scala/org/apache/spark/repl/SparkILoop.scala:54: variable addedClasspath in class ILoop is deprecated (since 2.11.0): use reset, replay or require to update class path
[warn] settings.classpath append addedClasspath
[warn] ^
[warn] two warnings found
[error] one error found
[error] Compile failed at Aug 29, 2018 5:28:22 PM [0.679s]
```
Readd the profile for `scala-2.12`. Using `-Pscala-2.12` will overrides `extra.source.dir` and `extra.testsource.dir` with two non-exist directories.
## How was this patch tested?
First, make sure it compiles.
```
dev/change-scala-version.sh 2.12
mvn -Pscala-2.12 -DskipTests compile install
```
Then, make a distribution to try the repl:
`./dev/make-distribution.sh --name custom-spark --tgz -Phadoop-2.7 -Phive -Pyarn -Pscala-2.12`
```
18/08/30 16:04:50 WARN NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
Using Spark's default log4j profile: org/apache/spark/log4j-defaults.properties
Setting default log level to "WARN".
To adjust logging level use sc.setLogLevel(newLevel). For SparkR, use setLogLevel(newLevel).
Spark context Web UI available at http://172.16.131.140:4040
Spark context available as 'sc' (master = local[*], app id = local-1535616298812).
Spark session available as 'spark'.
Welcome to
____ __
/ __/__ ___ _____/ /__
_\ \/ _ \/ _ `/ __/ '_/
/___/ .__/\_,_/_/ /_/\_\ version 2.4.0-SNAPSHOT
/_/
Using Scala version 2.12.6 (Java HotSpot(TM) 64-Bit Server VM, Java 1.8.0_112)
Type in expressions to have them evaluated.
Type :help for more information.
scala> spark.sql("select percentile(key, 1) from values (1, 1),(2, 1) T(key, value)").show
+-------------------------------------+
|percentile(key, CAST(1 AS DOUBLE), 1)|
+-------------------------------------+
| 2.0|
+-------------------------------------+
```
Closes#22280 from sadhen/SPARK_24785_FOLLOWUP.
Authored-by: 忍冬 <rendong@wacai.com>
Signed-off-by: Sean Owen <sean.owen@databricks.com>
## What changes were proposed in this pull request?
`JavaColumnExpressionSuite.java` was added and `org.apache.spark.sql.ColumnExpressionSuite#test("isInCollection: Java Collection")` was removed.
It provides native Java tests for the method `org.apache.spark.sql.Column#isInCollection`.
Closes#22253 from aai95/isInCollectionJavaTest.
Authored-by: aai95 <aai95@yandex.ru>
Signed-off-by: DB Tsai <d_tsai@apple.com>
After SPARK-18371, it is guaranteed that there would be at least one message per partition per batch using direct kafka API when new messages exist in the topics. This change will give the user the option of setting the minimum instead of just a hard coded 1 limit
The related unit test is updated and some internal tests verified that the topic partitions with new messages will be progressed by the specified minimum.
Author: Reza Safi <rezasafi@cloudera.com>
Closes#22223 from rezasafi/streaminglag.
## What changes were proposed in this pull request?
This PR is an follow-up PR of #21087 based on [a discussion thread](https://github.com/apache/spark/pull/21087#discussion_r211080067]. Since #21087 changed a condition of `if` statement, the message in an exception is not consistent of the current behavior.
This PR updates the exception message.
## How was this patch tested?
Existing UTs
Closes#22269 from kiszk/SPARK-23997-followup.
Authored-by: Kazuaki Ishizaki <ishizaki@jp.ibm.com>
Signed-off-by: Sean Owen <sean.owen@databricks.com>
## What changes were proposed in this pull request?
R tests require `testthat` v1.0.2. In the PR, I described how to install the version in the section http://spark.apache.org/docs/latest/building-spark.html#running-r-tests.
Closes#22272 from MaxGekk/r-testthat-doc.
Authored-by: Maxim Gekk <maxim.gekk@databricks.com>
Signed-off-by: hyukjinkwon <gurwls223@apache.org>
## What changes were proposed in this pull request?
Since https://github.com/apache/spark/pull/21696. Spark uses Parquet schema instead of Hive metastore schema to do pushdown.
That change can avoid wrong records returned when Hive metastore schema and parquet schema are in different letter cases. This pr add a test case for it.
More details:
https://issues.apache.org/jira/browse/SPARK-25206
## How was this patch tested?
unit tests
Closes#22267 from wangyum/SPARK-24716-TESTS.
Authored-by: Yuming Wang <yumwang@ebay.com>
Signed-off-by: Wenchen Fan <wenchen@databricks.com>
## What changes were proposed in this pull request?
Introduced by #21320 and #11744
```
$ sbt
> ++2.12.6
> project sql
> compile
...
[error] [warn] spark/sql/core/src/main/scala/org/apache/spark/sql/execution/ProjectionOverSchema.scala:41: match may not be exhaustive.
[error] It would fail on the following inputs: (_, ArrayType(_, _)), (_, _)
[error] [warn] getProjection(a.child).map(p => (p, p.dataType)).map {
[error] [warn]
[error] [warn] spark/sql/core/src/main/scala/org/apache/spark/sql/execution/ProjectionOverSchema.scala:52: match may not be exhaustive.
[error] It would fail on the following input: (_, _)
[error] [warn] getProjection(child).map(p => (p, p.dataType)).map {
[error] [warn]
...
```
And
```
$ sbt
> ++2.12.6
> project hive
> testOnly *ParquetMetastoreSuite
...
[error] /Users/rendong/wdi/spark/sql/hive/src/test/scala/org/apache/spark/sql/hive/HiveSparkSubmitSuite.scala:22: object tools is not a member of package scala
[error] import scala.tools.nsc.Properties
[error] ^
[error] /Users/rendong/wdi/spark/sql/hive/src/test/scala/org/apache/spark/sql/hive/HiveSparkSubmitSuite.scala:146: not found: value Properties
[error] val version = Properties.versionNumberString match {
[error] ^
[error] two errors found
...
```
## How was this patch tested?
Existing tests.
Closes#22260 from sadhen/fix_exhaustive_match.
Authored-by: 忍冬 <rendong@wacai.com>
Signed-off-by: hyukjinkwon <gurwls223@apache.org>
HyukjinKwon
## What changes were proposed in this pull request?
add __from pyspark.util import \_exception_message__ to python/pyspark/java_gateway.py
## How was this patch tested?
[flake8](http://flake8.pycqa.org) testing of https://github.com/apache/spark on Python 3.7.0
$ __flake8 . --count --select=E901,E999,F821,F822,F823 --show-source --statistics__
```
./python/pyspark/java_gateway.py:172:20: F821 undefined name '_exception_message'
emsg = _exception_message(e)
^
1 F821 undefined name '_exception_message'
1
```
Please review http://spark.apache.org/contributing.html before opening a pull request.
Closes#22265 from cclauss/patch-2.
Authored-by: cclauss <cclauss@bluewin.ch>
Signed-off-by: hyukjinkwon <gurwls223@apache.org>
Spark scheduler can hang when fetch failures, executor lost, task running on lost executor, and multiple stage attempts. To fix this we change to always unregister the pending partition on task completion.
## What changes were proposed in this pull request?
this PR is actually reverting the change in SPARK-19263, so that it always does shuffleStage.pendingPartitions -= task.partitionId. The change in SPARK-23433, should fix the issue originally from SPARK-19263.
## How was this patch tested?
Unit tests. The condition happens on a race which I haven't reproduced on a real customer, just see it sometimes on customers jobs in a real cluster.
I am also working on adding spark scheduler integration tests.
Closes#21976 from tgravescs/SPARK-24909.
Authored-by: Thomas Graves <tgraves@unharmedunarmed.corp.ne1.yahoo.com>
Signed-off-by: Marcelo Vanzin <vanzin@cloudera.com>
## What changes were proposed in this pull request?
`fromId` is the child, and `toId` is the parent, see line 127 in `buildSparkPlanGraphNode` above.
The edges in Spark UI also go from child to parent.
## How was this patch tested?
Comment change only. Inspected code above. Inspected how the edges in Spark UI look like.
Closes#22268 from juliuszsompolski/sparkplangraphedgedoc.
Authored-by: Juliusz Sompolski <julek@databricks.com>
Signed-off-by: Xiao Li <gatorsmile@gmail.com>
## What changes were proposed in this pull request?
I made one pass over the Python APIs for barrier mode and updated them to match the Scala doc in #22240 . Major changes:
* export the public classes
* expand the docs
* add doc for BarrierTaskInfo.addresss
cc: jiangxb1987
Closes#22261 from mengxr/SPARK-25248.1.
Authored-by: Xiangrui Meng <meng@databricks.com>
Signed-off-by: Xiangrui Meng <meng@databricks.com>
## What changes were proposed in this pull request?
BarrierCoordinator uses Timer and TimerTask. `TimerTask#cancel()` is invoked in ContextBarrierState#cancelTimerTask but `Timer#purge()` is never invoked.
Once a TimerTask is scheduled, the reference to it is not released until `Timer#purge()` is invoked even though `TimerTask#cancel()` is invoked.
## How was this patch tested?
I checked the number of instances related to the TimerTask using jmap.
Closes#22258 from sarutak/fix-barrierexec-oom.
Authored-by: sarutak <sarutak@oss.nttdata.co.jp>
Signed-off-by: Xiangrui Meng <meng@databricks.com>
## What changes were proposed in this pull request?
Alternative take on https://github.com/apache/spark/pull/22063 that does not introduce udfInternal.
Resolve issue with inferring func types in 2.12 by instead using info captured when UDF is registered -- capturing which types are nullable (i.e. not primitive)
## How was this patch tested?
Existing tests.
Closes#22259 from srowen/SPARK-25044.2.
Authored-by: Sean Owen <sean.owen@databricks.com>
Signed-off-by: Wenchen Fan <wenchen@databricks.com>
## What changes were proposed in this pull request?
This changes the calls of `toPandas()` and `createDataFrame()` to use the Arrow stream format, when Arrow is enabled. Previously, Arrow data was written to byte arrays where each chunk is an output of the Arrow file format. This was mainly due to constraints at the time, and caused some overhead by writing the schema/footer on each chunk of data and then having to read multiple Arrow file inputs and concat them together.
Using the Arrow stream format has improved these by increasing performance, lower memory overhead for the average case, and simplified the code. Here are the details of this change:
**toPandas()**
_Before:_
Spark internal rows are converted to Arrow file format, each group of records is a complete Arrow file which contains the schema and other metadata. Next a collect is done and an Array of Arrow files is the result. After that each Arrow file is sent to Python driver which then loads each file and concats them to a single Arrow DataFrame.
_After:_
Spark internal rows are converted to ArrowRecordBatches directly, which is the simplest Arrow component for IPC data transfers. The driver JVM then immediately starts serving data to Python as an Arrow stream, sending the schema first. It then starts a Spark job with a custom handler that sends Arrow RecordBatches to Python. Partitions arriving in order are sent immediately, and out-of-order partitions are buffered until the ones that precede it come in. This improves performance, simplifies memory usage on executors, and improves the average memory usage on the JVM driver. Since the order of partitions must be preserved, the worst case is that the first partition will be the last to arrive all data must be buffered in memory until then. This case is no worse that before when doing a full collect.
**createDataFrame()**
_Before:_
A Pandas DataFrame is split into parts and each part is made into an Arrow file. Then each file is prefixed by the buffer size and written to a temp file. The temp file is read and each Arrow file is parallelized as a byte array.
_After:_
A Pandas DataFrame is split into parts, then an Arrow stream is written to a temp file where each part is an ArrowRecordBatch. The temp file is read as a stream and the Arrow messages are examined. If the message is an ArrowRecordBatch, the data is saved as a byte array. After reading the file, each ArrowRecordBatch is parallelized as a byte array. This has slightly more processing than before because we must look each Arrow message to extract the record batches, but performance ends up a litle better. It is cleaner in the sense that IPC from Python to JVM is done over a single Arrow stream.
## How was this patch tested?
Added new unit tests for the additions to ArrowConverters in Scala, existing tests for Python.
## Performance Tests - toPandas
Tests run on a 4 node standalone cluster with 32 cores total, 14.04.1-Ubuntu and OpenJDK 8
measured wall clock time to execute `toPandas()` and took the average best time of 5 runs/5 loops each.
Test code
```python
df = spark.range(1 << 25, numPartitions=32).toDF("id").withColumn("x1", rand()).withColumn("x2", rand()).withColumn("x3", rand()).withColumn("x4", rand())
for i in range(5):
start = time.time()
_ = df.toPandas()
elapsed = time.time() - start
```
Current Master | This PR
---------------------|------------
5.803557 | 5.16207
5.409119 | 5.133671
5.493509 | 5.147513
5.433107 | 5.105243
5.488757 | 5.018685
Avg Master | Avg This PR
------------------|--------------
5.5256098 | 5.1134364
Speedup of **1.08060595**
## Performance Tests - createDataFrame
Tests run on a 4 node standalone cluster with 32 cores total, 14.04.1-Ubuntu and OpenJDK 8
measured wall clock time to execute `createDataFrame()` and get the first record. Took the average best time of 5 runs/5 loops each.
Test code
```python
def run():
pdf = pd.DataFrame(np.random.rand(10000000, 10))
spark.createDataFrame(pdf).first()
for i in range(6):
start = time.time()
run()
elapsed = time.time() - start
gc.collect()
print("Run %d: %f" % (i, elapsed))
```
Current Master | This PR
--------------------|----------
6.234608 | 5.665641
6.32144 | 5.3475
6.527859 | 5.370803
6.95089 | 5.479151
6.235046 | 5.529167
Avg Master | Avg This PR
---------------|----------------
6.4539686 | 5.4784524
Speedup of **1.178064192**
## Memory Improvements
**toPandas()**
The most significant improvement is reduction of the upper bound space complexity in the JVM driver. Before, the entire dataset was collected in the JVM first before sending it to Python. With this change, as soon as a partition is collected, the result handler immediately sends it to Python, so the upper bound is the size of the largest partition. Also, using the Arrow stream format is more efficient because the schema is written once per stream, followed by record batches. The schema is now only send from driver JVM to Python. Before, multiple Arrow file formats were used that each contained the schema. This duplicated schema was created in the executors, sent to the driver JVM, and then Python where all but the first one received are discarded.
I verified the upper bound limit by running a test that would collect data that would exceed the amount of driver JVM memory available. Using these settings on a standalone cluster:
```
spark.driver.memory 1g
spark.executor.memory 5g
spark.sql.execution.arrow.enabled true
spark.sql.execution.arrow.fallback.enabled false
spark.sql.execution.arrow.maxRecordsPerBatch 0
spark.driver.maxResultSize 2g
```
Test code:
```python
from pyspark.sql.functions import rand
df = spark.range(1 << 25, numPartitions=32).toDF("id").withColumn("x1", rand()).withColumn("x2", rand()).withColumn("x3", rand())
df.toPandas()
```
This makes total data size of 33554432×8×4 = 1073741824
With the current master, it fails with OOM but passes using this PR.
**createDataFrame()**
No significant change in memory except that using the stream format instead of separate file formats avoids duplicated the schema, similar to toPandas above. The process of reading the stream and parallelizing the batches does cause the record batch message metadata to be copied, but it's size is insignificant.
Closes#21546 from BryanCutler/arrow-toPandas-stream-SPARK-23030.
Authored-by: Bryan Cutler <cutlerb@gmail.com>
Signed-off-by: hyukjinkwon <gurwls223@apache.org>
## What changes were proposed in this pull request?
Using some reflection tricks to merge Scala 2.11 and 2.12 codebase.
## How was this patch tested?
Existing tests.
Closes#22246 from dbtsai/repl.
Lead-authored-by: DB Tsai <d_tsai@apple.com>
Co-authored-by: Liang-Chi Hsieh <viirya@gmail.com>
Signed-off-by: DB Tsai <d_tsai@apple.com>
This eliminates some duplication in the code to connect to a server on localhost to talk directly to the jvm. Also it gives consistent ipv6 and error handling. Two other incidental changes, that shouldn't matter:
1) python barrier tasks perform authentication immediately (rather than waiting for the BARRIER_FUNCTION indicator)
2) for `rdd._load_from_socket`, the timeout is only increased after authentication.
Closes#22247 from squito/py_connection_refactor.
Authored-by: Imran Rashid <irashid@cloudera.com>
Signed-off-by: hyukjinkwon <gurwls223@apache.org>
## What changes were proposed in this pull request?
`toAvroType` converts spark data type to avro schema. It always appends the record name to namespace so its impossible to have an Avro namespace independent of the record name.
When invoked with a spark data type like,
```java
val sparkSchema = StructType(Seq(
StructField("name", StringType, nullable = false),
StructField("address", StructType(Seq(
StructField("city", StringType, nullable = false),
StructField("state", StringType, nullable = false))),
nullable = false)))
// map it to an avro schema with record name "employee" and top level namespace "foo.bar",
val avroSchema = SchemaConverters.toAvroType(sparkSchema, false, "employee", "foo.bar")
// result is
// avroSchema.getName = employee
// avroSchema.getNamespace = foo.bar.employee
// avroSchema.getFullname = foo.bar.employee.employee
```
The patch proposes to fix this so that the result is
```
avroSchema.getName = employee
avroSchema.getNamespace = foo.bar
avroSchema.getFullname = foo.bar.employee
```
## How was this patch tested?
New and existing unit tests.
Please review http://spark.apache.org/contributing.html before opening a pull request.
Closes#22251 from arunmahadevan/avro-fix.
Authored-by: Arun Mahadevan <arunm@apache.org>
Signed-off-by: hyukjinkwon <gurwls223@apache.org>
