Career -

About

Designed and built one of Korea’s first large-scale distributed MySQL databases for a financial service, running tens of thousands of transactions per second across hundreds of production databases with zero downtime, while automating repetitive operations so that systems handle about 90% of routine work.

Later expanded focus from storing data to moving it — building and operating an in-house CDC platform that delivers large-scale change data from Oracle/MySQL/PostgreSQL to services in real time, along with a Kafka-based environment processing up to 1 million messages per second.

The goal is a data ecosystem that lets data flow reliably and freely to wherever it’s needed, without being tied to any specific technology, source, or target.

Highlights

Built a MySQL sharding data platform handling tens of thousands+ transactions per second
Operated hundreds of production databases with zero downtime (4 years)
Automated about 90% of routine database operations
Built a Kafka-based environment processing 1 million messages per second

Key Experience

KakaoBank Database Buildout

Designed and built KakaoBank’s early MySQL architecture
Designed a sharding structure handling tens of thousands+ transactions per second
Managed hundreds of production databases
Owned the full lifecycle: design, build, operations, incident response, performance tuning
Zero downtime since launch
Shared the buildout experience at external conferences

Database Operations Automation

Automated about 90% of routine DB operations
Systematized repetitive work — monitoring, alerting, routine checks
Built a Prometheus / Grafana-based monitoring environment
Reduced purely manual repetitive work to about 10%

CDC and Data Pipelines

Goal: sync data across heterogeneous Oracle / MySQL / PostgreSQL systems
Designed and developed an in-house CDC engine and platform
Applied and operated it in production (tens of thousands of events/sec)
Delivers DB change data to various services in real time
Built and operates a BI data pipeline connecting to an internal network

Kafka / Big Data Platform

Built a Kafka-based real-time data platform
Achieved throughput of 1 million messages per second
Built a real-time data pipeline connecting CDC and Kafka
Develops and operates the big data platform

Career

KakaoBank | Database / Data Platform

May 2016 ~ Present

Designed and operated large-scale distributed MySQL databases
DB operations automation and performance optimization
Developed and operated an in-house CDC platform
Built and operated a BI data pipeline
Built and operated a Kafka-based big data platform

Kakao | DBA

Dec 2012 ~ Apr 2016

Company-wide MySQL operations, incident response, and performance optimization
DB design review and technical support
Moved to the KakaoBank TF

TMON | DBA

Sep 2012 ~ Nov 2012

Company-wide MySQL operations, incident response, and performance optimization
DB design review

KTH (now KT alpha) | DBA

Jan 2008 ~ Sep 2012

Oracle / MySQL operations and performance optimization
Developed operations automation systems
Built and operated DB virtualization
Developed and operated the KT Megazone web service

OneStone Communication | Developer

Jan 2004 ~ Feb 2006

Developed and operated online ad servers
Developed and operated web services
Delivered enterprise SI projects

Talks

Books

MySQL Performance Optimization
MariaDB in Practice

Skills

MySQL / MariaDB / PostgreSQL / Oracle / CDC / Kafka / Java / Go / Docker / Prometheus / Grafana

Education

B.S. in Computer Engineering, Inha University