Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.
Wiki Markup
{scrollbar}{excerpt}How to use a PDI transformation to extract data from HBase and load it into a RDBMS table.{excerpt} The new RDBMS table will contain the count of page views by IP address and month.
{info:title=Note}For brevity's sake, this transformation will only contain three steps:  HBase Input, Split Fields and Table Output.    In practice, the full expressiveness of the PDI transformation semantic is available.  Further, PDI supports bulk loading for many RDBMS and that would be a viable, and common, alternative to using a Table Output approach.
{info}

h1. Prerequisites

In order to follow along with this how-to guide you will need the following:
* Hadoop
* Pentaho Data Integration
* HBase
* [Loading Data into HBase] guide completed

h1. Sample Files

There are no sample files for this guide.  The [Loading Data into HBase] guide must be completed prior to starting this guide as it loads the sample data.

h1. Step-By-Step Instructions


h2. Setup

Start Hadoop if it is not already running.

Start HBase if it is not already running.


{include:Create RDBMS Connection}

h2. Create a Transformation to Extract Data from HBase

In this task you will create a transformation to extract data from HBase and load into a RDBMS table.
{tip:title=Speed Tip}You can download the Kettle Transform [Extracting Data from HBase to Load an RDBMS in MapR^hbase_to_rdbms.ktr] already completed{tip}
# Start PDI on your desktop. Once it is running choose 'File' \-> 'New' \-> 'Transformation' from the menu system or click on the
'New file' icon on the toolbar and choose the 'Transformation' option.
\\
\\
# *Add a HBase Input Step:* You are going to extract data from HBase, so expand the 'Big Data' section of the Design palette and drag a 'HBase Input' node onto the transformation canvas.  Your transformation should look like:
!BAD:Common Images^AddHBaseInput.PNG|width=300,height=206!\\
\\
# *Edit the HBase Input* *Step*: Double-click on the 'HBase Input' node to edit its properties. Do the following:
## Zookeeper host(s) and Zookeeper port: Your HBase Zookeeper connection information.  For a local single node cluster your host is 'localhost' and your port is '2181'.
## HBase table name: Click 'Get mapped table names' and select 'weblogs'.
## Mapping name: Click 'Get mappings for the specified table' and select 'pageviews'.
## Click the 'Get Key/Fields Info' button to populate the grid
\\
When you are done your window should look like:
!EditHBaseInput.PNG|width=624,height=408!
Click 'OK' to close the window.
\\
\\
# *Add a Split Fields Step:* You need to split the key field which is client_ip\|year into two fields, so expand the 'Transform' section of the Design palette and drag a 'Split Fields' node onto the transformation canvas.  Your transformation should look like:
!BAD:Common Images^AddSplitFields.png|width=350,height=353!\\
# *Connect the Input and Split Fields steps*: Hover the mouse over the 'HBase Input' node and a tooltip will appear.  Click on the output connector (the green arrow pointing to the right) and drag a connector arrow to the 'Split Fields' node. Your canvas should look like this:
!BAD:Common Images^ConnectHBaseInputandSplitFields.png!\\
\\
# *Edit the Split Fields Step*: Double-click on the 'Split Fields' node to edit its properties. Do the following:
## Field to split: Select 'key'.
## Delimiter: Enter '\|'.
## Fields: Add the following:
| New field | Type |
| client_ip | String |
| year | Integer |
\\
When you are done your window should look like:
!BAD:Common Images^ConfigureSplitFields.png|width=297,height=194!
Click 'OK' to close the window.
\\
\\
# *Add a Table Output Step:* You want to write the values to a RDBMS, so expand the 'Output' section of the Design palette and drag a 'Table Output' node onto the transformation canvas.  Your transformation should look like:
!BAD:Common Images^AddTableOutputHBase.png|width=389,height=376!\\
\\
# *Connect the Split Fields and Table Output steps*: Hover the mouse over the 'Split Fields' node and a tooltip will appear. Click on the output connector (the green arrow pointing to the right) and drag a connector arrow to the 'Table Output' node. Your canvas should look like this:
!BAD:Common Images^ConnectSplitFieldsandTableOutput.png!\\
\\
# *Edit the Table Output* *Step*: Double-click on the 'Table Output' node to edit its properties. Do the following:
## Connection: Select 'RDBMS'
## Target Table: Enter 'aggregate_hbase'
## Check 'Truncate table' so you can re-run this transformation.
## Click the 'SQL' button to create the table in your target database.
!BAD:Common Images^HBaseSimpleSQLEditor.png|width=236,height=197!
## Click the 'Execute' button to run the SQL.
!BAD:Common Images^HBaseSQLResults.png|width=240,height=301!
The 'Results of the SQL statements' window will appear telling you if the SQL was successfully executed or give you troubleshooting information if the SQL failed.
## Click 'OK' to close the window.
## Click 'Close' to close the 'Simple SQL editor' window.
\\
When you are done your window should look like:
!BAD:Common Images^ConfigureTableOutputHBase.png!
Click 'OK' to close the window.
\\
\\
# *Save the Transformation*: Choose 'File' \-> 'Save as...' from the menu system. Save the transformation as 'hbase_to_rdbms.kjbktr' into a folder of your choice.
\\
\\
# *Run the Transformation*: Choose 'Action' \-> 'Run' from the menu system or click on the green run button on the transformation toolbar. A 'Execute a transformation' window will open. Click on the 'Launch' button. An 'Execution Results' panel will open at the bottom of the PDI window and it will show you the progress of the transformation as it runs. After several seconds the transformation should finish successfully: !worddav2afa26a775a81325a87a5451b1a9cd92.png|height=107,width=624!
If any errors occurred the job step that failed will be highlighted in red and you can use the 'Logging' tab to view error messages.
\\

h2. Check RDBMS for HBase Aggregated Table

# *Explore the Database:* Choose 'Tools' \-> 'Database' \-> 'Explore' from the menu system.
\\
\\
# *Select the Connection:* In the 'Make your selection' window select 'RDBMS' and click 'OK'.
!BAD:Common Images^SelectConnection.png!\\
\\
# *Preview the Table:* Expand RDBMS \-> Tables.  Right click on 'aggregate_hbase' and select 'Preview first 100'
!BAD:Common Images^PreviewTable.png!

h1. Summary

During this guide you learned how to use a PDI transformation to extract data from HBase and load into an RDBMS table.
{scrollbar}