CRUD operations in HBase in Hadoop - Mini Project: Build & Apply

CRUD operations in HBase

📖 Scenario: You work at a company that stores customer information in HBase, a big data database. You need to practice how to create, read, update, and delete data in HBase tables.

🎯 Goal: Build a simple HBase workflow to create a table, add data, read data, update data, and delete data using HBase shell commands.

📋 What You'll Learn

Use HBase shell commands to create a table

Insert data into the table

Read data from the table

Update existing data in the table

Delete data from the table

💡 Why This Matters

🌍 Real World

HBase is used in big data systems to store large amounts of data in a fast, scalable way. Knowing CRUD operations helps manage this data efficiently.

💼 Career

Many data engineer and big data analyst roles require working with HBase or similar NoSQL databases to store and retrieve data.

1

Create HBase table

In HBase shell, create a table called customers with one column family named info. Use the command create 'customers', 'info'.

Hadoop

# Type the HBase shell command to create the table
# Your code here

Hint

Use the create command with the table name and column family in single quotes.

2

Insert data into the table

Insert a row with row key row1 into the customers table. Add a column info:name with value John Doe using the command put 'customers', 'row1', 'info:name', 'John Doe'.

Hadoop

create 'customers', 'info'
# Type the HBase shell command to insert data
# Your code here

Hint

Use the put command with table name, row key, column, and value.

3

Read data from the table

Read the data for row key row1 from the customers table using the command get 'customers', 'row1'.

Hadoop

create 'customers', 'info'
put 'customers', 'row1', 'info:name', 'John Doe'
# Type the HBase shell command to read data
# Your code here

Hint

Use the get command with table name and row key to read data.

4

Update and delete data

Update the info:name column for row row1 to Jane Smith using put 'customers', 'row1', 'info:name', 'Jane Smith'. Then delete the row row1 using deleteall 'customers', 'row1'.

Hadoop

create 'customers', 'info'
put 'customers', 'row1', 'info:name', 'John Doe'
get 'customers', 'row1'
# Type the HBase shell commands to update and delete data
# Your code here

Hint

Use put to update and deleteall to delete the entire row.

Practice

1. Which HBase operation is used to retrieve data from a table?

easy

A. Get

B. Put

C. Delete

D. Scan

Solution

1. Step 1: Understand HBase CRUD operations

   HBase uses Get to read data, Put to create or update, Delete to remove, and Scan to read multiple rows.

2. Step 2: Identify the read operation

   Get is the specific operation to retrieve data for a single row.

3. Final Answer:

   Get -> Option A

4. Quick Check:

   Read = Get [OK]

Hint: Remember: Get = Read, Put = Create/Update [OK]

Common Mistakes:

• Confusing Put with Get
• Using Delete to read data
• Thinking Scan is for single row read

2. Which of the following is the correct syntax to insert data into HBase using Java API?

easy

A. Delete delete = new Delete(Bytes.toBytes("row1")); table.put(delete);

B. Get get = new Get(Bytes.toBytes("row1")); table.put(get);

C. Put put = new Put(Bytes.toBytes("row1")); put.addColumn(Bytes.toBytes("cf"), Bytes.toBytes("qual"), Bytes.toBytes("value")); table.put(put);

D. Put put = new Put("row1"); put.add("cf", "qual", "value"); table.get(put);

Solution

1. Step 1: Review correct Put syntax in HBase Java API

   Put requires a row key as bytes, then addColumn with column family, qualifier, and value as bytes, then call table.put(put).

2. Step 2: Check each option

   Put put = new Put(Bytes.toBytes("row1")); put.addColumn(Bytes.toBytes("cf"), Bytes.toBytes("qual"), Bytes.toBytes("value")); table.put(put); matches correct syntax. Get get = new Get(Bytes.toBytes("row1")); table.put(get); wrongly uses Get with put. Delete delete = new Delete(Bytes.toBytes("row1")); table.put(delete); uses Delete with put. Put put = new Put("row1"); put.add("cf", "qual", "value"); table.get(put); uses wrong constructors and methods.

3. Final Answer:

   Put put = new Put(Bytes.toBytes("row1")); put.addColumn(Bytes.toBytes("cf"), Bytes.toBytes("qual"), Bytes.toBytes("value")); table.put(put); -> Option C

4. Quick Check:

   Put syntax correct = Put put = new Put(Bytes.toBytes("row1")); put.addColumn(Bytes.toBytes("cf"), Bytes.toBytes("qual"), Bytes.toBytes("value")); table.put(put); [OK]

Hint: Put needs Bytes.toBytes for all keys and values [OK]

Common Mistakes:

• Using Get or Delete with put()
• Not converting strings to bytes
• Wrong method names like add() instead of addColumn()

3. What will be the output of the following HBase Java code snippet?

Put put = new Put(Bytes.toBytes("row1"));
put.addColumn(Bytes.toBytes("cf"), Bytes.toBytes("name"), Bytes.toBytes("Alice"));
table.put(put);

Get get = new Get(Bytes.toBytes("row1"));
Result result = table.get(get);
byte[] value = result.getValue(Bytes.toBytes("cf"), Bytes.toBytes("name"));
System.out.println(Bytes.toString(value));

medium

A. Compilation Error

B. null

C. cf:name

D. Alice

Solution

1. Step 1: Insert data with Put operation

   The code puts value "Alice" in column family "cf" and qualifier "name" for row "row1".

2. Step 2: Retrieve data with Get operation

   The Get fetches the row "row1" and extracts the value for "cf:name" which is "Alice".

3. Final Answer:

   Alice -> Option D

4. Quick Check:

   Put then Get returns inserted value [OK]

Hint: Put then Get returns inserted value exactly [OK]

Common Mistakes:

• Expecting null if not flushed
• Confusing column family and qualifier
• Forgetting to convert bytes to string

4. Identify the error in the following HBase delete operation code snippet:

Delete delete = new Delete("row1");
delete.addColumn(Bytes.toBytes("cf"), Bytes.toBytes("qual"));
table.delete(delete);

medium

A. Delete constructor argument should be bytes, not string

B. addColumn method is deprecated, should use addColumns

C. table.delete() method does not exist

D. No error, code is correct

Solution

1. Step 1: Check Delete constructor parameter type

   The Delete constructor requires a byte array for the row key, not a plain string.

2. Step 2: Verify method usage

   addColumn is valid for deleting a specific column. table.delete() is correct method.

3. Final Answer:

   Delete constructor argument should be bytes, not string -> Option A

4. Quick Check:

   Delete row key must be bytes [OK]

Hint: Always convert row keys to bytes in HBase operations [OK]

Common Mistakes:

• Passing string directly instead of bytes
• Confusing addColumn with addColumns
• Thinking table.delete() is invalid

5. You want to update the value of a column "status" in column family "info" for row "user123" in HBase. Which sequence of operations is correct to ensure the update is saved and can be read back immediately?

hard

A. Create a Delete for the old value, call table.delete(delete), then create a Put with new value and call table.put(put)

B. Create a Put with new value, call table.put(put), then call table.get(get) to read updated value

C. Call table.get(get) first, then create a Put with new value and call table.put(put)

D. Create a Put with new value, call table.get(get), then call table.put(put)

Solution

1. Step 1: Use Put to update the column value

   In HBase, Put operation overwrites existing values for the same column, so create a Put with the new "status" value for "user123".

2. Step 2: Save and verify update

   Call table.put(put) to save the update, then use table.get(get) to read back the updated value immediately.

3. Final Answer:

   Create a Put with new value, call table.put(put), then call table.get(get) to read updated value -> Option B

4. Quick Check:

   Put updates value, Get reads updated value [OK]

Hint: Put overwrites; Get reads latest value after put [OK]

Common Mistakes:

• Deleting before putting is unnecessary
• Getting before putting returns old value
• Calling get before put misses update