Snippets Groups Projects

[SPARK-4570][SQL]add BroadcastLeftSemiJoinHash

wangxiaojing authored 10 years ago

JIRA issue: [SPARK-4570](https://issues.apache.org/jira/browse/SPARK-4570)
We are planning to create a `BroadcastLeftSemiJoinHash` to implement the broadcast join for `left semijoin`
In left semijoin :
If the size of data from right side is smaller than the user-settable threshold `AUTO_BROADCASTJOIN_THRESHOLD`,
the planner would mark it as the `broadcast` relation and mark the other relation as the stream side. The broadcast table will be broadcasted to all of the executors involved in the join, as a `org.apache.spark.broadcast.Broadcast` object. It will use `joins.BroadcastLeftSemiJoinHash`.,else it will use `joins.LeftSemiJoinHash`.

The benchmark suggests these  made the optimized version 4x faster  when `left semijoin`
<pre><code>
Original:
left semi join : 9288 ms
Optimized:
left semi join : 1963 ms
</code></pre>
The micro benchmark load `data1/kv3.txt` into a normal Hive table.
Benchmark code:
<pre><code>
 def benchmark(f: => Unit) = {
    val begin = System.currentTimeMillis()
    f
    val end = System.currentTimeMillis()
    end - begin
  }
  val sc = new SparkContext(
    new SparkConf()
      .setMaster("local")
      .setAppName(getClass.getSimpleName.stripSuffix("$")))
  val hiveContext = new HiveContext(sc)
  import hiveContext._
  sql("drop table if exists left_table")
  sql("drop table if exists right_table")
  sql( """create table left_table (key int, value string)
       """.stripMargin)
  sql( s"""load data local inpath "/data1/kv3.txt" into table left_table""")
  sql( """create table right_table (key int, value string)
       """.stripMargin)
  sql(
    """
      |from left_table
      |insert overwrite table right_table
      |select left_table.key, left_table.value
    """.stripMargin)

  val leftSimeJoin = sql(
    """select a.key from left_table a
      |left semi join right_table b on a.key = b.key""".stripMargin)
  val leftSemiJoinDuration = benchmark(leftSimeJoin.count())
  println(s"left semi join : $leftSemiJoinDuration ms ")
</code></pre>

Author: wangxiaojing <u9jing@gmail.com>

Closes #3442 from wangxiaojing/SPARK-4570 and squashes the following commits:

a4a43c9 [wangxiaojing] rebase
f103983 [wangxiaojing] change style
fbe4887 [wangxiaojing] change style
ff2e618 [wangxiaojing] add testsuite
1a8da2a [wangxiaojing] add BroadcastLeftSemiJoinHash

07fa1910

07fa1910 10 years ago

Name	Last commit	Last update
assembly
bagel
bin
build
conf
core
data/mllib
dev
docker
docs
ec2
examples
external
extras
graphx
mllib
network
project
python
repl
sbin
sbt
sql
streaming
tools
yarn
.gitattributes
.gitignore
.rat-excludes
CONTRIBUTING.md
LICENSE
NOTICE
README.md
make-distribution.sh
pom.xml
scalastyle-config.xml
tox.ini

Apache Spark

Spark is a fast and general cluster computing system for Big Data. It provides high-level APIs in Scala, Java, and Python, 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 structured data processing, MLlib for machine learning, GraphX for graph processing, and Spark Streaming for stream processing.

http://spark.apache.org/

Online Documentation

You can find the latest Spark documentation, including a programming guide, on the project web page and project wiki. This README file only contains basic setup instructions.

Building Spark

Spark is built using Apache Maven. To build Spark and its example programs, run:

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 with Maven".

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 1000:

scala> sc.parallelize(1 to 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 1000:

>>> sc.parallelize(range(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-cluster" or "yarn-client" 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 all automated tests.

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" for detailed guidance on building for a particular distribution of Hadoop, including building for particular Hive and Hive Thriftserver distributions. See also "Third Party Hadoop Distributions" for guidance on building a Spark application that works with a particular distribution.

Configuration

Please refer to the Configuration guide in the online documentation for an overview on how to configure Spark.