Skip to content

hbase-standalone.md

Latest commit

 

History

History
182 lines (123 loc) · 5.03 KB

hbase-standalone.md

File metadata and controls

182 lines (123 loc) · 5.03 KB

HBase in Standalone Mode

1. Pull the required Docker Images and use them to Create and Run the Docker Containers

Delete the 4 previous containers then create and run the Docker containers specified in Docker-Compose.yaml i.e.,

  • HBase-Master
  • HBase-Regionserver
  • Zookeeper

2. Connect to the HBase Shell Hosted in the hbase-master Docker container

HBase Shell is a JRuby-based command-line program you can use to interact with HBase.

docker exec -it hbase-master hbase shell

You can also confirm that HBase is running via its Web-UI: http://localhost:16010/

Execute the following statements in HBase shell:

# To show the version of HBase
## The output according to the setup should be:
## version 2.1.3
version

# To show the details of the servers running HBase:
# The output according to the setup should be:
# 1 active master, 0 backup masters, 1 servers, 0 dead, 2.0000 average load
status

3. Getting Help in HBase

To get guidance on a specific command:

# Replace COMMAND with the command you want guidance on
help 'COMMAND'

For general guidance on how to use table-referenced commands.

table_help

4. Create a Table

The table product has 2 column families:

  • safety_features
  • custom_paint
create 'product', 'safety_features', 'custom_paint'

Verify that the table has been created:

list

5. View the Table's Metadata

Execute the following to view the metadata of the created table:

describe 'product'

6. Insert Data

We use the keyword put to insert data in HBase. The following statement inserts a new record with the key first adding present to the column 'blind_spot_monitoring' which is in the 'safety_features' column family.

put 'product', 'first', 'safety_features:blind_spot_monitoring', 'present'
put 'product', 'first', 'safety_features:parking_assist', 'present'

Unfortunately, the put command in HBase shell allows you to insert only one column value at a time.

7. Retrieve Data

We use the keyword get to retrieve data from HBase. get requires the table name and the row key.

get 'product', 'first', 'safety_features:blind_spot_monitoring'
get 'product', 'first', 'safety_features:parking_assist'

We use the keyword scan to retrieve all the rows. This is compute-intensive for large databases and should be avoided in production. By default, HBase uses the current timestamp when inserting data and the most recent timestamp when retrieving data.

scan 'product'

8. Altering Tables

Altering tables is computationally expensive because HBase creates a new column family with the chosen specifications and then copies all the data to the new column.

  • Disable the table
disable 'product'

By default, HBase stores only 3 versions of values (each with a timestamp). But this can be changed as follows:

alter 'product', { NAME => 'safety_features', VERSIONS => org.apache.hadoop.hbase.HConstants::ALL_VERSIONS }

We can also add a column-family (while the table is still disabled). The new column family called make.

alter 'product', { NAME => 'make', VERSIONS => org.apache.hadoop.hbase.HConstants::ALL_VERSIONS }

Similar to the safety_features column family, the make column family is added without any columns. It is upon the user to honour the schema. However, if the user decides not to honour the schema, e.g., by adding data to make:new_column, HBase will not stop them.

Lastly, we can set the compression method as follows:

alter 'product', {NAME=>'safety_features', COMPRESSION=>'GZ', BLOOMFILTER=>'ROW'}
  • Enable the table
enable 'product'

9. Insert more Data with Schema-Flexibility in the make Column-Family

put 'product', 'second', 'safety_features:blind_spot_monitoring', 'present'
put 'product', 'second', 'safety_features:adaptive_cruise_control', 'present'
put 'product', 'second', 'safety_features:lane_keeping_assist', 'present'
scan 'product'
put 'product', 'second', 'custom_paint:red', '96'
put 'product', 'second', 'custom_paint:blue', '96'
put 'product', 'second', 'custom_paint:green', '96'
put 'product', 'first', 'make:YOM_brand', '2014 Mazda CX6'

put 'product', 'second', 'make:year_of_manufacture', '2019'
put 'product', 'second', 'make:brand', 'Toyota RAV4'
scan 'product'

10. Update Existing Data

HBase does not have a separate "update" command because put automatically overwrites any existing data in the specified row and column.

put 'product', 'first', 'safety_features:parking_assist', 'not present'
put 'product', 'second', 'custom_paint:blue', '100'

11. Delete Existing Data

HBase does not have a separate "update" command because put automatically overwrites any existing data in the specified row and column.

delete 'product', 'first', 'safety_features:parking_assist'
delete 'product', 'second', 'custom_paint:green'