From 59ab83c4038970c2a642f401df397da1b7c61f96 Mon Sep 17 00:00:00 2001
From: heibaiying <31504331+heibaiying@users.noreply.github.com>
Date: Sat, 27 Apr 2019 10:32:25 +0800
Subject: [PATCH] =?UTF-8?q?Update=20Hive=E5=88=86=E5=8C=BA=E8=A1=A8?=
=?UTF-8?q?=E5=92=8C=E5=88=86=E6=A1=B6=E8=A1=A8.md?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
---
notes/Hive分区表和分桶表.md | 7 ++-----
1 file changed, 2 insertions(+), 5 deletions(-)
diff --git a/notes/Hive分区表和分桶表.md b/notes/Hive分区表和分桶表.md
index e9f898b..8c18b23 100644
--- a/notes/Hive分区表和分桶表.md
+++ b/notes/Hive分区表和分桶表.md
@@ -79,7 +79,7 @@ LOAD DATA LOCAL INPATH "/usr/file/emp30.txt" OVERWRITE INTO TABLE emp_partition
在HashMap中，当我们给put()方法传递键和值时，我们先对键调用hashCode()方法，返回的hashCode用于找到bucket(桶)位置，最后将键值对存储在对应桶的链表结构中，链表达到一定阈值后会转换为红黑树(JDK1.8+)。下图为HashMap的数据结构图：
-
+
> 图片引用自：[HashMap vs. Hashtable](http://www.itcuties.com/java/hashmap-hashtable/)
@@ -113,9 +113,6 @@ LOAD DATA LOCAL INPATH "/usr/file/emp30.txt" OVERWRITE INTO TABLE emp_partition
```sql
set hive.enforce.bucketing = true; --Hive 2.x不需要这一步
```
-
-需要在插入分桶的时候hash, **也就是说向分桶表中插入数据的时候必然要执行一次MAPREDUCE,**
-
在Hive 0.x and 1.x版本，必须使用设置`hive.enforce.bucketing = true`，表示强制分桶，允许程序根据表结构自动选择正确数量的Reducer和cluster by column来进行分桶。
#### 2. CTAS导入数据
@@ -168,4 +165,4 @@ SELECT * FROM page_view WHERE dt='2009-02-25';
## 参考资料
-1. [LanguageManual DDL BucketedTables](https://cwiki.apache.org/confluence/display/Hive/LanguageManual+DDL+BucketedTables)
\ No newline at end of file
+1. [LanguageManual DDL BucketedTables](https://cwiki.apache.org/confluence/display/Hive/LanguageManual+DDL+BucketedTables)
