A comprehensive Docker-based solution for MariaDB performance testing with integrated Percona PMM monitoring.

├── docker-compose.yml     # Complete environment setup
├── Dockerfile            # Test application container
├── README.md             # This documentation
└── test_files/           # Test scripts directory
    ├── auto_run_tests.sh     # Automated test runner
    ├── setup_database.sh     # Database setup
    ├── insert_test.sh        # INSERT operations
    ├── update_test.sh        # UPDATE operations  
    ├── delete_test.sh        # DELETE operations
    ├── select_test.sh        # SELECT queries
    └── run_performance_tests.sh  # Interactive test menu

# Start all services
docker compose up -d

# Enable query response time plugin
./mariadb_setup/enable-query-response-time.sh

# Setup PMM table statistics  
./mariadb_setup/enable-pmm-table-stats.sh

# Run performance tests on custom database
sudo docker run --rm -e DB_NAME=sales_db --name sales_db_test --network mariadb_app_network mariadb-tester

• Container: mariadb1
• Port: 3306
• Credentials: root / 123
• Default DB: performance_test

• Container: pmm-server
• Web UI: http://localhost:8080
• HTTPS UI: https://localhost:8443
• Login: admin / admin

• Container: pmm-client
• Purpose: Monitoring agent for MariaDB
• Auto-connects: Links to PMM server automatically

• Built from: Dockerfile
• Purpose: CRUD and performance testing
• Database Parameter: Configurable via DB_NAME environment variable
• Usage: Run separately using docker commands

# First, build the test image
docker build -t mariadb-tester .

# Run tests on default database (performance_test)
sudo docker run --rm --network mariadb_app_network mariadb-tester

# Run tests on custom database (simplified syntax)
sudo docker run --rm --network mariadb_app_network \
  -e DB_NAME=sales_db --name sales_db_test mariadb-tester

# Run interactive test menu
sudo docker run --rm -it --network mariadb_app_network \
  -e DB_NAME=sales_db --name sales_db_test mariadb-tester ./run_performance_tests.sh

1. Start services: docker compose up -d
2. Open browser: http://localhost:8080
3. Login: admin / admin

� Setup PMM Monitoring:

# Enable query response time plugin
./mariadb_setup/enable-query-response-time.sh

# Setup PMM table statistics (follow instructions)
./mariadb_setup/enable-pmm-table-stats.sh

Manual Registration via Web UI:

1. Go to Configuration → Inventory
2. Click "Add Service"
3. Select "MySQL"
4. Use these connection details:
   • Host: mariadb1
   • Port: 3306
   • Username: root
   • Password: 123
   • Service Name: mariadb-monitoring
5. ⚙️ CRITICAL - Advanced Settings:
   • ✅ Enable "Table Statistics"
   • ✅ Set "Table Statistics Limit" to 5000
   • ✅ Enable "Query Analytics" with "Performance Schema"
   • ✅ Enable "User Statistics"

All test scripts support these parameters:

DB_NAME=your_database    # Target database name
DB_HOST=mariadb1         # Database host
DB_PORT=3306             # Database port  
DB_USER=root             # Database username
DB_PASS=123              # Database password

# Build test image first (if not already built)
sudo docker build -t mariadb-tester .

# Test specific database (simplified)
sudo docker run --rm --network mariadb_app_network \
  -e DB_NAME=sales_db --name sales_db_test mariadb-tester

# Test with different database
sudo docker run --rm --network mariadb_app_network \
  -e DB_NAME=inventory --name inventory_test mariadb-tester

# Test with custom user credentials (override defaults)
sudo docker run --rm --network mariadb_app_network \
  -e DB_NAME=inventory -e DB_USER=testuser -e DB_PASS=testpass \
  --name inventory_test mariadb-tester

# Start core services (MariaDB + PMM)
docker compose up -d

# Stop all services  
docker compose down

# View logs
docker compose logs -f mariadb1
docker compose logs -f pmm-server

# Clean restart
docker compose down -v && docker compose up -d

# Build test application
sudo docker build -t mariadb-tester .

# Run specific test
sudo docker run --rm --network mariadb_app_network \
  -e DB_NAME=sales_db --name sales_db_test \
  mariadb-tester ./insert_test.sh

CREATE TABLE transactions (
    transaction_id INT AUTO_INCREMENT PRIMARY KEY,
    product_name VARCHAR(100),
    quantity INT,
    price DECIMAL(10,2),
    transaction_date DATETIME
);

• Dockerized Environment - Complete setup with one command
• Database Flexibility - Test any database via DB_NAME parameter
• Comprehensive Testing - CRUD operations + performance tests
• Real-time Monitoring - Percona PMM integration
• Network Isolation - All services on dedicated network
• No Auto-registration - Manual PMM configuration for control

# 1. Start core services
docker compose up -d

# 2. Build test application
sudo docker build -t mariadb-tester .

# 3. Run tests on your database  
sudo docker run --rm --network mariadb_app_network \
  -e DB_NAME=myapp_db --name myapp_test mariadb-tester

# 4. Monitor in PMM dashboard
# Visit: http://localhost:8080

# 5. Clean up
docker compose down

This setup provides a complete MariaDB testing and monitoring environment with minimal configuration required.

DB_NAME="performance_test"
DB_USER="root"
DB_PASS=""  # Set your password
DB_HOST="localhost"
DB_PORT="3306"

cd /home/magdi/mariadb
chmod +x *.sh
./run_performance_tests.sh

# Run all tests
./run_performance_tests.sh all

# Run individual tests
./run_performance_tests.sh setup
./run_performance_tests.sh insert
./run_performance_tests.sh update
./run_performance_tests.sh delete
./run_performance_tests.sh select

# Show statistics
./run_performance_tests.sh stats

# Clean database
./run_performance_tests.sh clean

# Setup database first
./setup_database.sh

# Run performance tests
./insert_test.sh      # Insert 100,000 records
./select_test.sh      # Various SELECT patterns
./update_test.sh      # Update 50,000 records
./delete_test.sh      # Delete 25,000 records

# Optimize MariaDB configuration
./optimize_mariadb.sh

• Records: 100,000 (configurable)
• Batch Size: 1,000 records per batch
• Threads: 4 parallel threads
• Data: Realistic product transactions with random data
• Features: Bulk inserts with transaction batching for optimal performance

• Updates: 50,000 (configurable)
• Patterns: Price adjustments, quantity updates, product name changes, date updates
• Threads: 4 parallel threads
• Features: Various update patterns to simulate real-world scenarios

• Deletes: 25,000 (configurable)
• Patterns: Old records, low-value transactions, high quantities, specific products
• Threads: 4 parallel threads
• Features: Different deletion criteria to test various index usage

• Query Types: Simple, Aggregate, Complex
• Queries: 1,000 per type (configurable)
• Threads: 4 parallel threads
• Features:
  • Simple: ID ranges, product filters, price ranges
  • Aggregate: COUNT, SUM, AVG, GROUP BY operations
  • Complex: Subqueries, window functions, statistical analysis

The optimizer script automatically:

• Detects system resources (RAM, CPU, disk space)
• Calculates optimal buffer pool size (70% of RAM)
• Sets appropriate connection limits
• Optimizes InnoDB settings for bulk operations
• Disables safety features for maximum performance (testing only!)

• InnoDB Buffer Pool: 70% of available RAM
• Log File Size: 25% of buffer pool size
• Flush Settings: Optimized for performance over durability
• Thread Pool: Based on CPU cores
• Disabled Features: Binary logging, doublewrite buffer (testing only)

• INSERT: 15,000-25,000 records/second
• SELECT Simple: 5,000-10,000 queries/second
• SELECT Aggregate: 1,000-3,000 queries/second
• SELECT Complex: 500-1,500 queries/second
• UPDATE: 5,000-15,000 updates/second
• DELETE: 3,000-10,000 deletes/second

Performance varies significantly based on hardware, MariaDB configuration, and system load.

Each test script provides:

• Real-time progress updates
• Execution time measurement
• Throughput calculations (operations/second)
• Error reporting
• Sample result verification

INSERT test completed!
Total time: 6.45s
Records inserted: 100000
Throughput: 15503.88 records/second
Current table count: 100000 records

⚠️ WARNING: These scripts are designed for performance testing only!

• Scripts disable important safety features (durability, binary logging)
• Use dedicated test systems, not production databases
• High-volume operations may impact system performance
• Ensure adequate disk space before running tests

❌ "No data" in PMM Table Details Dashboard:

# Enable query response time
./mariadb_setup/enable-query-response-time.sh

# Setup PMM table statistics
./mariadb_setup/enable-pmm-table-stats.sh

🔧 Manual Fix:

1. Ensure Table Statistics are enabled during service registration
2. Set Table Statistics Limit to 5000+ (not default 1000)
3. Enable Performance Schema for Query Analytics
4. Wait 2-3 minutes after registration for data collection

1. Connection Failed

   • Check MariaDB service status: sudo docker compose ps
   • Verify credentials in script configuration
   • Ensure MariaDB is accepting connections

2. Low Performance

   • Check system resources (RAM, CPU, I/O)
   • Monitor system load during tests
   • Use docker compose logs mariadb1 for debugging

3. Out of Memory Errors

   • Reduce batch sizes in scripts
   • Decrease thread count
   • Increase system swap space
   • Optimize MariaDB memory settings

4. Permission Errors

   • Ensure scripts are executable: chmod +x *.sh
   • Use sudo for Docker commands
   • Verify file system permissions

Edit variables in each script:

TOTAL_RECORDS=100000    # Number of records to insert
BATCH_SIZE=1000         # Records per batch
THREADS=4               # Parallel threads
QUERIES_PER_TYPE=1000   # Queries per test type

Modify the test functions in each script to add:

• Different data patterns
• Custom query types
• Additional product categories
• Specific business logic scenarios

1. Run baseline tests with default MariaDB settings
2. Apply optimizations with optimize_mariadb.sh
3. Run tests again and compare results
4. Adjust individual parameters and re-test

# Create a performance monitoring loop
while true; do
    echo "Starting performance test cycle at $(date)"
    ./run_performance_tests.sh all
    echo "Sleeping for 1 hour..."
    sleep 3600
done

For issues or questions:

1. Check MariaDB error logs: /var/log/mysql/error.log
2. Monitor system resources: htop, iotop, free -h
3. Review MariaDB performance schema
4. Adjust script parameters based on your system capabilities

Note: This test suite is designed for database performance evaluation and should be used in controlled test environments only.