d50f8df929
### What changes were proposed in this pull request? This PR implements a tiny performance optimization for a `GenericArrayData` constructor, avoiding an unnecessary roundtrip through `WrappedArray` when the provided value is already an array of objects. It also fixes a related performance problem in `ParquetRowConverter`. ### Why are the changes needed? `GenericArrayData` has a `this(seqOrArray: Any)` constructor, which was originally added in #13138 for use in `RowEncoder` (where we may not know concrete types until runtime) but is also called (perhaps unintentionally) in a few other code paths. In this constructor's existing implementation, a call to `new WrappedArray(Array[Object](""))` is dispatched to the `this(seqOrArray: Any)` constructor, where we then call `this(array.toSeq)`: this wraps the provided array into a `WrappedArray`, which is subsequently unwrapped in a `this(seq.toArray)` call. For an interactive example, see https://scastie.scala-lang.org/7jOHydbNTaGSU677FWA8nA This PR changes the `this(seqOrArray: Any)` constructor so that it calls the primary `this(array: Array[Any])` constructor, allowing us to save a `.toSeq.toArray` call; this comes at the cost of one additional `case` in the `match` statement (but I believe this has a negligible performance impact relative to the other savings). As code cleanup, I also reverted the JVM 1.7 workaround from #14271. I also fixed a related performance problem in `ParquetRowConverter`: previously, this code called `ArrayBasedMapData.apply` which, in turn, called the `this(Any)` constructor for `GenericArrayData`: this PR's micro-benchmarks show that this is _significantly_ slower than calling the `this(Array[Any])` constructor (and I also observed time spent here during other Parquet scan benchmarking work). To fix this performance problem, I replaced the call to the `ArrayBasedMapData.apply` method with direct calls to the `ArrayBasedMapData` and `GenericArrayData` constructors. ### Does this PR introduce any user-facing change? No. ### How was this patch tested? I tested this by running code in a debugger and by running microbenchmarks (which I've added to a new `GenericArrayDataBenchmark` in this PR): - With JDK8 benchmarks: this PR's changes more than double the performance of calls to the `this(Any)` constructor. Even after improvements, however, calls to the `this(Array[Any])` constructor are still ~60x faster than calls to `this(Any)` when passing a non-primitive array (thereby motivating this patch's other change in `ParquetRowConverter`). - With JDK11 benchmarks: the changes more-or-less completely eliminate the performance penalty associated with the `this(Any)` constructor. Closes #27088 from JoshRosen/joshrosen/GenericArrayData-optimization. Authored-by: Josh Rosen <rosenville@gmail.com> Signed-off-by: Dongjoon Hyun <dhyun@apple.com> |
||
---|---|---|
.github | ||
assembly | ||
bin | ||
build | ||
common | ||
conf | ||
core | ||
data | ||
dev | ||
docs | ||
examples | ||
external | ||
graphx | ||
hadoop-cloud | ||
launcher | ||
licenses | ||
licenses-binary | ||
mllib | ||
mllib-local | ||
project | ||
python | ||
R | ||
repl | ||
resource-managers | ||
sbin | ||
sql | ||
streaming | ||
tools | ||
.gitattributes | ||
.gitignore | ||
appveyor.yml | ||
CONTRIBUTING.md | ||
LICENSE | ||
LICENSE-binary | ||
NOTICE | ||
NOTICE-binary | ||
pom.xml | ||
README.md | ||
scalastyle-config.xml |
Apache Spark
Spark is a unified analytics engine for large-scale data processing. It provides high-level APIs in Scala, Java, Python, and R, and an optimized engine that supports general computation graphs for data analysis. It also supports a rich set of higher-level tools including Spark SQL for SQL and DataFrames, MLlib for machine learning, GraphX for graph processing, and Structured Streaming for stream processing.
Online Documentation
You can find the latest Spark documentation, including a programming guide, on the project web page. This README file only contains basic setup instructions.
Building Spark
Spark is built using Apache Maven. To build Spark and its example programs, run:
./build/mvn -DskipTests clean package
(You do not need to do this if you downloaded a pre-built package.)
More detailed documentation is available from the project site, at "Building Spark".
For general development tips, including info on developing Spark using an IDE, see "Useful Developer Tools".
Interactive Scala Shell
The easiest way to start using Spark is through the Scala shell:
./bin/spark-shell
Try the following command, which should return 1,000,000,000:
scala> spark.range(1000 * 1000 * 1000).count()
Interactive Python Shell
Alternatively, if you prefer Python, you can use the Python shell:
./bin/pyspark
And run the following command, which should also return 1,000,000,000:
>>> spark.range(1000 * 1000 * 1000).count()
Example Programs
Spark also comes with several sample programs in the examples
directory.
To run one of them, use ./bin/run-example <class> [params]
. For example:
./bin/run-example SparkPi
will run the Pi example locally.
You can set the MASTER environment variable when running examples to submit
examples to a cluster. This can be a mesos:// or spark:// URL,
"yarn" to run on YARN, and "local" to run
locally with one thread, or "local[N]" to run locally with N threads. You
can also use an abbreviated class name if the class is in the examples
package. For instance:
MASTER=spark://host:7077 ./bin/run-example SparkPi
Many of the example programs print usage help if no params are given.
Running Tests
Testing first requires building Spark. Once Spark is built, tests can be run using:
./dev/run-tests
Please see the guidance on how to run tests for a module, or individual tests.
There is also a Kubernetes integration test, see resource-managers/kubernetes/integration-tests/README.md
A Note About Hadoop Versions
Spark uses the Hadoop core library to talk to HDFS and other Hadoop-supported storage systems. Because the protocols have changed in different versions of Hadoop, you must build Spark against the same version that your cluster runs.
Please refer to the build documentation at "Specifying the Hadoop Version and Enabling YARN" for detailed guidance on building for a particular distribution of Hadoop, including building for particular Hive and Hive Thriftserver distributions.
Configuration
Please refer to the Configuration Guide in the online documentation for an overview on how to configure Spark.
Contributing
Please review the Contribution to Spark guide for information on how to get started contributing to the project.