Hive partitioned EXTERNAL tables, create, load data, select, managing

First of all there is creation of database for studing external tables. Before, look on directory structure


$ hadoop fs -ls /user/data
Found 6 items
-rw-r--r--   2 hadoop supergroup        161 2018-03-02 14:31 /user/data/cc2.avsc
drwxr-xr-x   - hadoop supergroup          0 2018-04-03 12:27 /user/data/dics
drwxr-xr-x   - hadoop supergroup          0 2018-03-15 15:42 /user/data/js1.json
drwxr-xr-x   - hadoop supergroup          0 2018-03-16 10:52 /user/data/js_db
drwxr-xr-x   - hadoop supergroup          0 2018-03-02 13:49 /user/data/order.parquet
-rw-r--r--   2 dr.who supergroup        389 2018-02-28 16:10 /user/data/test_data.csv

Create new one database with set location folder and setting some properties, like comment and creator. 

CREATE DATABASE IF NOT EXISTS ext_tabs 
COMMENT 'Database for studing external tables'
LOCATION '/user/data/exttabs'
WITH DBPROPERTIES ('creator' = 'Yakushev Aleksey', 'date' = '2018-04-05');

Now check database

DESCRIBE DATABASE ext_tabs;

db_name  |comment                              |location                                    |owner_name |owner_type |parameters |
---------|-------------------------------------|--------------------------------------------|-----------|-----------|-----------|
ext_tabs |Database for studing external tables |hdfs://hadoop-master:9000/user/data/exttabs |hadoop     |USER       |           |

Here we create our first external table

/*
 SET  - Stock Exchange of Thailand 
 NYSE - New York Stock Exchange
*/
DROP TABLE IF EXISTS ext_tabs.stocks;

CREATE EXTERNAL TABLE ext_tabs.stocks(
symbol STRING,
ts     TIMESTAMP,
price  FLOAT
)
PARTITIONED BY (year INT,exch STRING) 
ROW FORMAT SERDE 'org.openx.data.jsonserde.JsonSerDe'
STORED AS TEXTFILE;

And create table partitions: 

alter table ext_tabs.stocks add 
partition (year=2017,exch='SET') 
partition (year=2018,exch='SET')
partition (year=2017,exch='NYSE')
partition (year=2018,exch='NYSE');

Look at the directory structure 

$ hadoop fs -ls /user/data/exttabs/stocks
Found 2 items
drwxr-xr-x   - hadoop supergroup          0 2018-04-05 11:02 /user/data/exttabs/stocks/year=2017
drwxr-xr-x   - hadoop supergroup          0 2018-04-05 11:02 /user/data/exttabs/stocks/year=2018

$ hadoop fs -ls /user/data/exttabs/stocks/year=2017
Found 2 items
drwxr-xr-x   - hadoop supergroup          0 2018-04-05 11:02 /user/data/exttabs/stocks/year=2017/exch=NYSE
drwxr-xr-x   - hadoop supergroup          0 2018-04-05 11:02 /user/data/exttabs/stocks/year=2017/exch=SET

Create 2 json file with test data and load it into HDFS

-- -- load data into one partition with one json file into /user/data/exttabs/stocks/year=2017/exch=SET
-- 2017_set_part_1.json
{"symbol":"ORCL","ts":"2017-04-05 11:34:12.120000000","price":12.34}
{"symbol":"ORCL","ts":"2017-04-05 11:34:12.240000000","price":13.34}
{"symbol":"ORCL","ts":"2017-04-05 11:34:13.370000000","price":14.34}
 
-- -- load data into one partition with one json file into /user/data/exttabs/stocks/year=2018/exch=NYSE
-- 2018_nyse_part_1.json
{"symbol":"IBM","ts":"2018-04-05 11:34:12.220000000","price":22.34}
{"symbol":"IBM","ts":"2018-04-05 11:34:12.340000000","price":23.34}
{"symbol":"IBM","ts":"2018-04-05 11:34:13.470000000","price":24.34}

Check the data of table. 

select 
 stocks.symbol,
 cast(stocks.ts as double),
 stocks.ts,
 stocks.price,
 stocks.year,
 stocks.exch 
from 
 ext_tabs.stocks
 
stocks.symbol |ts            |stocks.ts           |stocks.price |stocks.year |stocks.exch |
--------------|--------------|--------------------|-------------|------------|------------|
ORCL          |1491381252.12 |2017-04-05 11:34:12 |12.34        |2017        |SET         |
ORCL          |1491381252.24 |2017-04-05 11:34:12 |13.34        |2017        |SET         |
ORCL          |1491381253.37 |2017-04-05 11:34:13 |14.34        |2017        |SET         |
IBM           |1522917252.22 |2018-04-05 11:34:12 |22.34        |2018        |NYSE        |
IBM           |1522917252.34 |2018-04-05 11:34:12 |23.34        |2018        |NYSE        |
IBM           |1522917253.47 |2018-04-05 11:34:13 |24.34        |2018        |NYSE        |

Additionally I can say thar Hive's external table doesn't store real data, it's just metadata. When you drop table the data will not erased and you can later recreate table. External table it's like a mask for data. 

-- This script can drop partition 
alter table ext_tabs.stocks drop partition (year=2019, exch='LSE')

insert into ext_tabs.stocks partition(year=2019, exch='LSE')
select 
 'AMZ' as symbol,
 cast(stocks.ts as double) as ts,
 stocks.price
from 
 ext_tabs.stocks

Hive automatically create directory structure for new PARTITION (Hive feature -dynamic partition )

$ hadoop fs -ls /user/data/exttabs/stocks
Found 3 items
drwxr-xr-x   - hadoop supergroup          0 2018-04-05 11:02 /user/data/exttabs/stocks/year=2017
drwxr-xr-x   - hadoop supergroup          0 2018-04-05 11:02 /user/data/exttabs/stocks/year=2018
drwxr-xr-x   - hadoop supergroup          0 2018-04-06 08:21 /user/data/exttabs/stocks/year=2019

Select new data with HQL

select * from ext_tabs.stocks where year=2019

stocks.symbol |stocks.ts           |stocks.price |stocks.year |stocks.exch |
--------------|--------------------|-------------|------------|------------|
AMZ           |2017-04-05 08:34:12 |12.34        |2019        |LSE         |
AMZ           |2017-04-05 08:34:12 |13.34        |2019        |LSE         |
AMZ           |2017-04-05 08:34:13 |14.34        |2019        |LSE         |
AMZ           |2018-04-05 08:34:12 |22.34        |2019        |LSE         |
AMZ           |2018-04-05 08:34:12 |23.34        |2019        |LSE         |
AMZ           |2018-04-05 08:34:13 |24.34        |2019        |LSE         |

File with new data is created (we can download it and open in text editor).

$ hadoop fs -ls /user/data/exttabs/stocks/year=2019/exch=LSE
Found 1 items
-rwxr-xr-x   1 hadoop supergroup        354 2018-04-06 08:22 /user/data/exttabs/stocks/year=2019/exch=LSE/000000_0

{"symbol":"AMZ","price":12.34,"ts":"2017-04-05T08:34:12Z"}
{"symbol":"AMZ","price":13.34,"ts":"2017-04-05T08:34:12Z"}
{"symbol":"AMZ","price":14.34,"ts":"2017-04-05T08:34:13Z"}
{"symbol":"AMZ","price":22.34,"ts":"2018-04-05T08:34:12Z"}
{"symbol":"AMZ","price":23.34,"ts":"2018-04-05T08:34:12Z"}
{"symbol":"AMZ","price":24.34,"ts":"2018-04-05T08:34:13Z"}

It's also easy export data from tables:
$ hadoop fs -mkdir /user/data/exttabs/export

insert overwrite directory 'hdfs://hadoop-master:9000/user/data/exttabs/export'
select * 
from ext_tabs.stocks 
where year=2017;

$ hadoop fs -ls /user/data/exttabs/export
Found 1 items
-rwxr-xr-x   1 hadoop supergroup        129 2018-04-09 09:43 /user/data/exttabs/export/000000_0

If there were more reducers writing output, we would have additional files with names e.g. 000001_0, 000002_0


Комментарии

Отправить комментарий

Популярные сообщения из этого блога

Loading data into Spark from Oracle RDBMS, CSV

In this article shown how read data from Oracle tables with jdbc and direct from csv with Spark. First of all we need download jdbc Oracle driver  ojdbc6.jar and put it into Spark Home jar directory. For example - C:\spark-2.3.0-bin-hadoop2.7\jars Example #1. From Oracle partitioned table load data per pertitions into parquet file. aaa.xxx.yyyy.zzz - ip address of Oracle server. import java.io.File import org.apache.spark.sql.{Row, SaveMode, SparkSession} val warehouseLocation = new File("spark-warehouse").getAbsolutePath val spark = SparkSession.builder().appName("Spark Hive Example").config("spark.sql.warehouse.dir",warehouseLocation).config("hive.exec.dynamic.partition","true").config("hive.exec.dynamic.partition.mode","nonstrict").enableHiveSupport().getOrCreate() //recreate table spark.sql(s"DROP TABLE IF EXISTS T_DATA") spark.sql(s"CREATE TABLE IF NOT EXISTS T_DATA(id_row STRING,
Далее...

Load data from Cassandra to HDFS parquet files and select with Hive

Изображение
HDFS directory preparation on Hadoop cluster. From LXC with name hdpnn execute next: Hadoop and Cassandra cluster installation you can find  in this article. My idea is writing an application with Scala which will be run on Spark cluster for load data from Cassandra into HDFS parquet files, for future analyzes with Hive. First I do it manually, step by step. 1) HDFS directory preparation [hadoop@hdpnn ~]$ hadoop fs -mkdir /user/tickers [hadoop@hdpnn ~]$ hadoop fs -rm -r /user/tickers/ticker_23 [hadoop@hdpnn ~]$ hdfs dfs -chown root:root /user/tickers [hadoop@hdpnn ~]$ hadoop fs -ls /user drwxr-xr-x - root root 0 2019-01-10 06:27 /user/tickers 2) Load data with Spark using spark-shell (on smn) spark-shell --driver-memory 1G --executor-memory 1G --driver-cores 1 --executor-cores 1 --jars "/opt/spark-2.3.2/jars/spark-cassandra-connector-assembly-2.3.2.jar" --conf spark.cassandra.connection.host= 192.168.122.192 --conf "spark.sql.parquet.writeL
Далее...

Hadoop 3.0 cluster - installation, configuration, tests on Cent OS 7

Изображение
In this article presented step by step how create Hadoop cluster with 1 name node and 3 slaves. At first point there is one VM (from hardware administrator) with properties: 10.242.5.88 root/123456Qw # cat /etc/centos-release CentOS Linux release 7.1.1503 (Core) # uname -a Linux ders-hadoop1 3.10.0-229.el7.x86_64 #1 SMP Fri Mar 6 11:36:42 UTC 2015 x86_64 x86_64 x86_64 GNU/Linux Storage information: Файловая система Размер Использовано Дост Использовано% Cмонтировано в /dev/mapper/centos-root 50G 914M 50G 2% / devtmpfs 1,9G 0 1,9G 0% /dev tmpfs 1,9G 0 1,9G 0% /dev/shm tmpfs 1,9G 8,3M 1,9G 1% /run tmpfs 1,9G 0 1,9G 0% /sys/fs/cgroup /dev/mapper/centos-home 73G 33M 73G 1% /home /dev/sda1 497M 11
Далее...