3.4-3.6 Hive Storage Format

一、file format

ORCFile在HDP 2:更好的压缩，更好的性能：

https://zh.hortonworks.com/blog/orcfile-in-hdp-2-better-compression-better-performance/

官方ORCfile介绍：

https://cwiki.apache.org/confluence/display/Hive/LanguageManual+ORC

官方PARQUET介绍：

https://cwiki.apache.org/confluence/display/Hive/Parquet

file_format:
  : SEQUENCEFILE
  | TEXTFILE    -- (Default, depending on hive.default.fileformat configuration)
  | RCFILE      -- (Note: Available in Hive 0.6.0 and later)
  | ORC         -- (Note: Available in Hive 0.11.0 and later)
  | PARQUET     -- (Note: Available in Hive 0.13.0 and later)
  | AVRO        -- (Note: Available in Hive 0.14.0 and later)
  | JSONFILE    -- (Note: Available in Hive 4.0.0 and later)
  | INPUTFORMAT input_format_classname OUTPUTFORMAT output_format_classname

数据存储：
    按行存储
    按列存储

#TEXTFILE ：行式存储
#其他的都是列存储，ORC是对RCFILE的优化；ORC和PARQUET用的较多；

二、测试

我们建三张表导入同样的数据，看数据存储的大小；

1、TEXTFILE

#建表
hive (default)> create table page_views(
              > track_time string,
              > url string,
              > session_id string,
              > referer string,
              > ip string,
              > end_user_id string,
              > city_id string
              > )
              > ROW FORMAT DELIMITED FIELDS TERMINATED BY '\t'
              > STORED AS TEXTFILE;
OK
Time taken: 0.089 seconds

#加载数据
hive (default)> load data local inpath '/opt/datas/page_views.data' into table page_views;
Copying data from file:/opt/datas/page_views.data
Copying file: file:/opt/datas/page_views.data
Loading data to table default.page_views
Table default.page_views stats: [numFiles=1, numRows=0, totalSize=19014993, rawDataSize=0]
OK
Time taken: 0.326 seconds

2、ORCFILE

hive (default)> create table page_views_orc(
              > track_time string,
              > url string,
              > session_id string,
              > referer string,
              > ip string,
              > end_user_id string,
              > city_id string
              > )
              > ROW FORMAT DELIMITED FIELDS TERMINATED BY '\t'
              > STORED AS orc;
OK
Time taken: 0.04 seconds

hive (default)> insert into table page_views_orc select * from page_views;

3、PARQUET

hive (default)> create table page_views_parquet(
              > track_time string,
              > url string,
              > session_id string,
              > referer string,
              > ip string,
              > end_user_id string,
              > city_id string
              > )
              > ROW FORMAT DELIMITED FIELDS TERMINATED BY '\t'
              > STORED AS parquet;
OK
Time taken: 0.037 seconds

hive (default)> insert into table page_views_parquet select * from page_views;

4、查看各个表文件大小

hive (default)> dfs -du -h /user/hive/warehouse/page_views/;
18.1 M  /user/hive/warehouse/page_views/page_views.data

hive (default)> dfs -du -h /user/hive/warehouse/page_views_orc/;
2.6 M  /user/hive/warehouse/page_views_orc/000000_0

hive (default)> dfs -du -h /user/hive/warehouse/page_views_parquet/;
13.1 M  /user/hive/warehouse/page_views_parquet/000000_0

#明显的可以看出，数据相同时，存储为不同的格式：
TEXTFILE格式的表：18.1M
ORCFILE格式的表：2.6M
PARQUET格式的表：13.1M

5、查询测试

#TEXTFILE表查询
hive (default)> select session_id, count(*) cnt from page_views group by session_id order by cnt desc limit 30;
......
Time taken: 39.427 seconds, Fetched: 30 row(s)        #39.427秒

#ORCFILE表查询
hive (default)> select session_id, count(*) cnt from page_views_orc group by session_id order by cnt desc limit 30;
......
Time taken: 38.319 seconds, Fetched: 30 row(s)        #38.319秒

#可见orc格式的表，不仅数据体积小很多，查询也有优势

三、ORCFILE优化

创建ORC表时，可以使用许多表属性进一步优化ORC的工作方式：

1、例如，假设您想使用snappy压缩而不是zlib压缩。方法如下:

#建表
hive (default)> create table page_views_orc_snappy(
              > track_time string,
              > url string,
              > session_id string,
              > referer string,
              > ip string,
              > end_user_id string,
              > city_id string
              > )
              > ROW FORMAT DELIMITED FIELDS TERMINATED BY '\t'
              > STORED AS orc tblproperties ("orc.compress"="SNAPPY");
OK
Time taken: 0.033 seconds

#插入数据
hive (default)> insert into table page_views_orc_snappy select * from page_views;

#查询
hive (default)> dfs -du -h /user/hive/warehouse/page_views_orc_snappy/;
3.8 M  /user/hive/warehouse/page_views_orc_snappy/000000_0

#此时发现，用snappy压缩完，数据反而比orcfile表数据还大了，怎么回事呢？
因为建ORCFILE表时，不指定压缩，默认使用ZLIB压缩，但是指定了snappy压缩就会使用snappy，
而且ZLIB的压缩比大于snappy，所以此时的snappy压缩稍微大一些；

总结：

在实际的项目开发当中，hive表的数据
    *存储格式I orcfile/qarquet    #推荐
    *数据压缩
        snappy        #推荐