ODIn/spark-instrumented-optimizer
1
0
Fork 0
Code Issues Pull requests Projects Releases Wiki Activity
spark-instrumented-optimizer/sql
History
Cheng Lian 8f7141fbc0 [SPARK-1368][SQL] Optimized HiveTableScan 
JIRA issue: [SPARK-1368](https://issues.apache.org/jira/browse/SPARK-1368)

This PR introduces two major updates:

- Replaced FP style code with `while` loop and reusable `GenericMutableRow` object in critical path of `HiveTableScan`.
- Using `ColumnProjectionUtils` to help optimizing RCFile and ORC column pruning.

My quick micro benchmark suggests these two optimizations made the optimized version 2x and 2.5x faster when scanning CSV table and RCFile table respectively:

```
Original:

[info] CSV: 27676 ms, RCFile: 26415 ms
[info] CSV: 27703 ms, RCFile: 26029 ms
[info] CSV: 27511 ms, RCFile: 25962 ms

Optimized:

[info] CSV: 13820 ms, RCFile: 10402 ms
[info] CSV: 14158 ms, RCFile: 10691 ms
[info] CSV: 13606 ms, RCFile: 10346 ms
```

The micro benchmark loads a 609MB CVS file (structurally similar to the `src` test table) into a normal Hive table with `LazySimpleSerDe` and a RCFile table, then scans these tables respectively.

Preparation code:

```scala
package org.apache.spark.examples.sql.hive

import org.apache.spark.sql.hive.LocalHiveContext
import org.apache.spark.{SparkConf, SparkContext}

object HiveTableScanPrepare extends App {
  val sparkContext = new SparkContext(
    new SparkConf()
      .setMaster("local")
      .setAppName(getClass.getSimpleName.stripSuffix("$")))

  val hiveContext = new LocalHiveContext(sparkContext)

  import hiveContext._

  hql("drop table scan_csv")
  hql("drop table scan_rcfile")

  hql("""create table scan_csv (key int, value string)
        |  row format serde 'org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe'
        |  with serdeproperties ('field.delim'=',')
      """.stripMargin)

  hql(s"""load data local inpath "${args(0)}" into table scan_csv""")

  hql("""create table scan_rcfile (key int, value string)
        |  row format serde 'org.apache.hadoop.hive.serde2.columnar.ColumnarSerDe'
        |stored as
        |  inputformat 'org.apache.hadoop.hive.ql.io.RCFileInputFormat'
        |  outputformat 'org.apache.hadoop.hive.ql.io.RCFileOutputFormat'
      """.stripMargin)

  hql(
    """
      |from scan_csv
      |insert overwrite table scan_rcfile
      |select scan_csv.key, scan_csv.value
    """.stripMargin)
}
```

Benchmark code:

```scala
package org.apache.spark.examples.sql.hive

import org.apache.spark.sql.hive.LocalHiveContext
import org.apache.spark.{SparkConf, SparkContext}

object HiveTableScanBenchmark extends App {
  val sparkContext = new SparkContext(
    new SparkConf()
      .setMaster("local")
      .setAppName(getClass.getSimpleName.stripSuffix("$")))

  val hiveContext = new LocalHiveContext(sparkContext)

  import hiveContext._

  val scanCsv = hql("select key from scan_csv")
  val scanRcfile = hql("select key from scan_rcfile")

  val csvDuration = benchmark(scanCsv.count())
  val rcfileDuration = benchmark(scanRcfile.count())

  println(s"CSV: $csvDuration ms, RCFile: $rcfileDuration ms")

  def benchmark(f: => Unit) = {
    val begin = System.currentTimeMillis()
    f
    val end = System.currentTimeMillis()
    end - begin
  }
}
```

@marmbrus Please help review, thanks!

Author: Cheng Lian <lian.cs.zju@gmail.com>

Closes #758 from liancheng/fastHiveTableScan and squashes the following commits:

4241a19 [Cheng Lian] Distinguishes sorted and possibly not sorted operations more accurately in HiveComparisonTest
cf640d8 [Cheng Lian] More HiveTableScan optimisations:
bf0e7dc [Cheng Lian] Added SortedOperation pattern to match *some* definitely sorted operations and avoid some sorting cost in HiveComparisonTest.
6d1c642 [Cheng Lian] Using ColumnProjectionUtils to optimise RCFile and ORC column pruning
eb62fd3 [Cheng Lian] [SPARK-1368] Optimized HiveTableScan
2014-05-29 15:24:03 -07:00
..
catalyst [SPARK-1938] [SQL] ApproxCountDistinctMergeFunction should return Int value. 2014-05-27 22:17:50 -07:00
core [SPARK-1368][SQL] Optimized HiveTableScan 2014-05-29 15:24:03 -07:00
hive [SPARK-1368][SQL] Optimized HiveTableScan 2014-05-29 15:24:03 -07:00
README.md [SPARK-1342] Scala 2.10.4 2014-04-01 18:35:50 -07:00

README.md

Spark SQL

This module provides support for executing relational queries expressed in either SQL or a LINQ-like Scala DSL.

Spark SQL is broken up into three subprojects:

  • Catalyst (sql/catalyst) - An implementation-agnostic framework for manipulating trees of relational operators and expressions.
  • Execution (sql/core) - A query planner / execution engine for translating Catalyst’s logical query plans into Spark RDDs. This component also includes a new public interface, SQLContext, that allows users to execute SQL or LINQ statements against existing RDDs and Parquet files.
  • Hive Support (sql/hive) - Includes an extension of SQLContext called HiveContext that allows users to write queries using a subset of HiveQL and access data from a Hive Metastore using Hive SerDes. There are also wrappers that allows users to run queries that include Hive UDFs, UDAFs, and UDTFs.

Other dependencies for developers

In order to create new hive test cases , you will need to set several environmental variables.

export HIVE_HOME="<path to>/hive/build/dist"
export HIVE_DEV_HOME="<path to>/hive/"
export HADOOP_HOME="<path to>/hadoop-1.0.4"

Using the console

An interactive scala console can be invoked by running sbt/sbt hive/console. From here you can execute queries and inspect the various stages of query optimization.

catalyst$ sbt/sbt hive/console

[info] Starting scala interpreter...
import org.apache.spark.sql.catalyst.analysis._
import org.apache.spark.sql.catalyst.dsl._
import org.apache.spark.sql.catalyst.errors._
import org.apache.spark.sql.catalyst.expressions._
import org.apache.spark.sql.catalyst.plans.logical._
import org.apache.spark.sql.catalyst.rules._
import org.apache.spark.sql.catalyst.types._
import org.apache.spark.sql.catalyst.util._
import org.apache.spark.sql.execution
import org.apache.spark.sql.hive._
import org.apache.spark.sql.hive.TestHive._
Welcome to Scala version 2.10.4 (Java HotSpot(TM) 64-Bit Server VM, Java 1.7.0_45).
Type in expressions to have them evaluated.
Type :help for more information.

scala> val query = sql("SELECT * FROM (SELECT * FROM src) a")
query: org.apache.spark.sql.ExecutedQuery =
SELECT * FROM (SELECT * FROM src) a
=== Query Plan ===
Project [key#6:0.0,value#7:0.1]
 HiveTableScan [key#6,value#7], (MetastoreRelation default, src, None), None

Query results are RDDs and can be operated as such.

scala> query.collect()
res8: Array[org.apache.spark.sql.execution.Row] = Array([238,val_238], [86,val_86], [311,val_311]...

You can also build further queries on top of these RDDs using the query DSL.

scala> query.where('key === 100).toRdd.collect()
res11: Array[org.apache.spark.sql.execution.Row] = Array([100,val_100], [100,val_100])

From the console you can even write rules that transform query plans. For example, the above query has redundant project operators that aren't doing anything. This redundancy can be eliminated using the transform function that is available on all TreeNode objects.

scala> query.logicalPlan
res1: catalyst.plans.logical.LogicalPlan = 
Project {key#0,value#1}
 Project {key#0,value#1}
  MetastoreRelation default, src, None


scala> query.logicalPlan transform {
     |   case Project(projectList, child) if projectList == child.output => child
     | }
res2: catalyst.plans.logical.LogicalPlan = 
Project {key#0,value#1}
 MetastoreRelation default, src, None