50일 차 – MySQL에서 ClickHouse ® 로 데이터를 마이그레이션하는 방법: 단계별 가이드

Introduction

As applications grow, traditional relational databases such as MySQL may struggle with analytical workloads involving millions of records and complex aggregations. While MySQL excels at Online Transaction Processing (OLTP), ClickHouse® is purpose-built for Online Analytical Processing (OLAP), enabling lightning-fast analytical queries on massive datasets.

Migrating data from MySQL to ClickHouse® allows organizations to build high-performance reporting systems, dashboards, and real-time analytics without impacting transactional workloads.

In this guide, you’ll learn several approaches to migrate data from MySQL to ClickHouse®, along with their advantages, limitations, and ideal use cases.

Why Migrate from MySQL to ClickHouse®?

MySQL and ClickHouse® are designed for different workloads.

Feature MySQL ClickHouse® Storage Model Row-based Columnar Best For Transactions (OLTP) Analytics (OLAP) Query Speed Fast for row lookups Extremely fast for large scans Aggregation Performance Moderate Extremely fast Scalability Primarily Vertical Optimized for analytical scaling Typical Use Cases Applications and transactional systems Reporting, dashboards, and analytics

Migrating from MySQL to ClickHouse® makes sense when:

• Analytical queries are becoming slow in MySQL.
• You need real-time dashboards over large datasets.
• Reporting queries are impacting your production database.
• You regularly process millions or billions of rows.

Migration Architecture

MySQL
   │
   ▼
Export / Synchronization
   │
   ▼
Data Transformation
   │
   ▼
ClickHouse®
   │
   ▼
Dashboards / Analytics

Enter fullscreen mode Exit fullscreen mode

Migration Methods

There are multiple ways to migrate data depending on your requirements.

Method 1: CSV Export and Import (Recommended for Beginners)

This is the simplest approach for performing a one-time migration of historical data.

Step 1: Export Data from MySQL

Run the following command inside MySQL:

SELECT *
INTO OUTFILE '/tmp/employees.csv'
FIELDS TERMINATED BY ','
ENCLOSED BY '"'
LINES TERMINATED BY '\n'
FROM employees;

Enter fullscreen mode Exit fullscreen mode

Alternatively, export using the MySQL CLI:

mysql -u root -p mydb \
-e "SELECT * FROM orders" \
| sed 's/\t/,/g' > /tmp/orders.csv

Enter fullscreen mode Exit fullscreen mode

Step 2: Create the Destination Table in ClickHouse®

Before importing data, create the corresponding MergeTree table. Remember to map MySQL data types to ClickHouse® equivalents.

CREATE TABLE employees
(
    id UInt32,
    name String,
    department String,
    salary Float64
)
ENGINE = MergeTree
ORDER BY id;

Enter fullscreen mode Exit fullscreen mode

Step 3: Import the CSV File

Using clickhouse-client:

clickhouse-client \
--query "INSERT INTO employees FORMAT CSV" \
< /tmp/employees.csv

Enter fullscreen mode Exit fullscreen mode

Using the HTTP API:

curl -u default:password \
"http://localhost:8123/?query=INSERT+INTO+default.employees+FORMAT+CSV" \
--data-binary @/tmp/employees.csv

Enter fullscreen mode Exit fullscreen mode

Step 4: Verify the Import

SELECT count()
FROM default.employees;

Enter fullscreen mode Exit fullscreen mode

Example output:

Method 2: Using the MySQL Table Engine

ClickHouse® can directly connect to and query MySQL tables without requiring an intermediate export.

Step 1: Create a MySQL Engine Table

CREATE TABLE mysql_employees
(
    id UInt32,
    name String,
    department String,
    salary Float64
)
ENGINE = MySQL(
    'localhost:3306',
    'default',
    'employees',
    'user',
    'password'
);

Enter fullscreen mode Exit fullscreen mode

You can now query the MySQL table directly.

SELECT *
FROM mysql_employees;

Enter fullscreen mode Exit fullscreen mode

Step 2: Create the Destination MergeTree Table

CREATE TABLE employees
(
    id UInt32,
    name String,
    department String,
    salary Float64
)
ENGINE = MergeTree
ORDER BY id;

Enter fullscreen mode Exit fullscreen mode

Step 3: Copy the Data

Once the MySQL engine is configured, migrate the data using a single query.

INSERT INTO employees
SELECT *
FROM mysql_employees;

Enter fullscreen mode Exit fullscreen mode

This is one of the fastest and cleanest approaches for one-time migrations.

Step 4: Verify the Migration

SELECT *
FROM employees
LIMIT 5;

Enter fullscreen mode Exit fullscreen mode

For most one-time migrations, this is the recommended approach because everything happens inside ClickHouse® without intermediate files.

Method 3: Using ClickPipes (ClickHouse Cloud)

If you’re using ClickHouse Cloud, ClickPipes provides a fully managed ingestion service.

Migration steps:

1. Create a ClickPipe.
2. Connect your MySQL database.
3. Select the tables you want to synchronize.
4. Start continuous replication.

Advantages

• Fully managed service
• Continuous incremental synchronization
• Minimal operational maintenance

Limitation

• Available only for ClickHouse Cloud.

Method 4: Real-Time Migration with Kafka and Debezium

For continuous synchronization between MySQL and ClickHouse®, Change Data Capture (CDC) using Debezium is the most robust solution.

Debezium captures every INSERT, UPDATE, and DELETE operation from MySQL and streams the changes into Kafka, where ClickHouse® consumes them.

Architecture

MySQL
   │
   ▼
Debezium (CDC)
   │
   ▼
Kafka
   │
   ▼
ClickHouse Kafka Engine
   │
   ▼
Materialized View
   │
   ▼
MergeTree Table

Enter fullscreen mode Exit fullscreen mode

This architecture enables near real-time synchronization.

Best suited for:

• Event streaming
• Real-time dashboards
• IoT platforms
• Monitoring systems
• Operational analytics

MySQL to ClickHouse® Data Type Mapping

MySQL Type ClickHouse® Type INT Int32 / UInt32 BIGINT Int64 / UInt64 VARCHAR(n) String TEXT String DECIMAL(p,s) Decimal(p,s) FLOAT Float32 DOUBLE Float64 DATETIME DateTime DATE Date TINYINT(1) UInt8 (Boolean) JSON String

Verifying the Migration

Always verify that both databases contain identical data after migration.

Row Count Validation

MySQL:

SELECT COUNT(*)
FROM orders;

Enter fullscreen mode Exit fullscreen mode

ClickHouse®:

SELECT count()
FROM default.orders;

Enter fullscreen mode Exit fullscreen mode

Both queries should return identical counts.

If they don’t, investigate:

• Data type mismatches
• NULL handling differences
• Missing records during export
• Failed import batches

Best Practices

1. Choose the Right ORDER BY Key

Unlike MySQL’s primary key, ClickHouse® uses ORDER BY to determine how data is physically sorted.

Choose columns that are frequently used in filtering conditions.

2. Use LowCardinality for Repetitive Strings

Columns with relatively few distinct values benefit from better compression.

country LowCardinality(String),
status LowCardinality(String)

Enter fullscreen mode Exit fullscreen mode

3. Partition by Time

For time-series data, partition by month or day.

PARTITION BY toYYYYMM(order_date)

Enter fullscreen mode Exit fullscreen mode

4. Remove AUTO_INCREMENT

ClickHouse® doesn’t support AUTO_INCREMENT.

Instead, simply store identifiers as:

UInt32

Enter fullscreen mode Exit fullscreen mode

or

UInt64

Enter fullscreen mode Exit fullscreen mode

5. Handle NULL Values Carefully

ClickHouse® columns are NOT NULL by default.

If nullable columns exist in MySQL, either use Nullable(Type) or replace NULL values during migration.

Example:

SELECT
    id,
    COALESCE(department, 'Unknown') AS department
FROM default.mysql_employee;

Enter fullscreen mode Exit fullscreen mode

Common Challenges

Data Type Mismatches

Ensure MySQL data types are correctly mapped before migration.

Duplicate Records

Implement deduplication strategies or unique identifiers when performing incremental loads.

Large Tables

Split large datasets into smaller batches to reduce memory consumption and improve reliability.

Character Encoding

Use UTF-8 encoding consistently to prevent text corruption.

Performance Tips

• Use the MergeTree engine for analytical workloads.
• Batch inserts instead of inserting one row at a time.
• Compress data whenever possible.
• Choose an efficient partitioning strategy.
• Avoid unnecessary indexes—ClickHouse® relies on sorting and data-skipping indexes instead.

Quick Reference

Scenario Recommended Method One-time migration CSV Export & Import Direct access to MySQL MySQL Table Engine Full table migration INSERT INTO ... SELECT Continuous synchronization ClickPipes Real-time streaming Kafka + Debezium

Conclusion

Migrating data from MySQL to ClickHouse® is an effective way to accelerate analytical workloads while keeping your transactional database unchanged.

In this architecture, MySQL continues handling application writes and transactional operations, while ClickHouse® powers dashboards, reporting, and large-scale analytical queries.

For most organizations, using the MySQL Table Engine together with INSERT INTO ... SELECT offers the simplest and most reliable solution for one-time migrations.

If continuous synchronization is required, Debezium with Kafka provides a production-ready Change Data Capture (CDC) pipeline capable of replicating changes in near real time.

With proper schema design, efficient partitioning, and the right migration strategy, ClickHouse® can deliver dramatically faster analytical performance from the very first day.