### What changes were proposed in this pull request? Document DISTRIBUTE BY clause of SELECT statement in SQL Reference Guide. ### Why are the changes needed? Currently Spark lacks documentation on the supported SQL constructs causing confusion among users who sometimes have to look at the code to understand the usage. This is aimed at addressing this issue. ### Does this PR introduce any user-facing change? Yes. **Before:** There was no documentation for this. **After.** <img width="972" alt="Screen Shot 2020-01-20 at 3 08 24 PM" src="https://user-images.githubusercontent.com/14225158/72763045-c08fbc80-3b96-11ea-8fb6-023cba5eb96a.png"> <img width="972" alt="Screen Shot 2020-01-20 at 3 08 34 PM" src="https://user-images.githubusercontent.com/14225158/72763047-c38aad00-3b96-11ea-80d8-cd3d2d4257c8.png"> ### How was this patch tested? Tested using jykyll build --serve Closes #27298 from dilipbiswal/sql-ref-select-distributeby. Authored-by: Dilip Biswal <dkbiswal@gmail.com> Signed-off-by: Sean Owen <srowen@gmail.com>
2.6 KiB
layout | title | displayTitle | license |
---|---|---|---|
global | DISTRIBUTE BY Clause | DISTRIBUTE BY Clause | Licensed to the Apache Software Foundation (ASF) under one or more contributor license agreements. See the NOTICE file distributed with this work for additional information regarding copyright ownership. The ASF licenses this file to You under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at http://www.apache.org/licenses/LICENSE-2.0 Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License. |
The DISTRIBUTE BY
clause is used to repartition the data based
on the input expressions. Unlike the CLUSTER BY
clause, this does not
sort the data within each partition.
Syntax
{% highlight sql %} DISTRIBUTE BY { expression [ , ... ] } {% endhighlight %}
Parameters
expression
- Specifies combination of one or more values, operators and SQL functions that results in a value.
Examples
{% highlight sql %} CREATE TABLE person (name STRING, age INT); INSERT INTO person VALUES ('Zen Hui', 25), ('Anil B', 18), ('Shone S', 16), ('Mike A', 25), ('John A', 18), ('Jack N', 16);
-- Reduce the number of shuffle partitions to 2 to illustrate the behavior of DISTRIBUTE BY
.
-- It's easier to see the clustering and sorting behavior with less number of partitions.
SET spark.sql.shuffle.partitions = 2;
-- Select the rows with no ordering. Please note that without any sort directive, the result
-- of the query is not deterministic. It's included here to just contrast it with the
-- behavior of DISTRIBUTE BY
. The query below produces rows where age columns are not
-- clustered together.
SELECT age, name FROM person;
+---+-------+ |age|name | +---+-------+ |16 |Shone S| |25 |Zen Hui| |16 |Jack N | |25 |Mike A | |18 |John A | |18 |Anil B | +---+-------+
-- Produces rows clustered by age. Persons with same age are clustered together.
-- Unlike CLUSTER BY
clause, the rows are not sorted within a partition.
SELECT age, name FROM person DISTRIBUTE BY age;
+---+-------+ |age|name | +---+-------+ |25 |Zen Hui| |25 |Mike A | |18 |John A | |18 |Anil B | |16 |Shone S| |16 |Jack N | +---+-------+ {% endhighlight %}