KAI

KAI Improves Railway Efficiencies with IoT Platform
using MySQL Enterprise Edition

"MySQL Enterprise Edition performance is sky high. We have full stability, no hung database or downtime, and smooth railway operations."

Husni Rijal,
Sr DBA, PT Kereta Api Indonesia (Persero)

National railways operator of Indonesia boosts database performance by 70% and cuts backup time by over 90% with MySQL Enterprise Edition.

Business Challenges & Goals

PT Kereta Api Indonesia (KAI) manages the national railways of Indonesia, operating 7,000 kilometers of rail networks in Java and Sumatra. The state company's rolling stock includes 420 diesel locomotives, 1000 multiple-unit trains, 1,700 passenger coaches, and 6,400 freight cars.

In 2019, KAI transported 429 million passengers and 47 million tons of cargo by rail. The company's medium-term plans include networks to be built on the currently railway-free islands of Kalimantan and Sulawesi.

For the safety of passengers, in 2014, KAI implemented a locomotive speed detection application named LocoTrack. The Internet of Things (IoT) platform uses on-board monitoring devices interconnected with outdoor signaling systems that can regulate train speeds and command trains to slow down.

Initially, KAI used the MySQL Community Edition database for running LocoTrack and for its rail cargo system. With increasing volumes of data generated from the IoT system, it was not getting enough performance, backup was taking too long, and downtime was compromising drivers and passenger safety.

Business Results & Metrics

In 2018, KAI upgraded to MySQL Enterprise Edition. Performance of the LocoTrack application increased by 70%, thanks to MySQL Enterprise Edition features such as MySQL indexing for reduced query execution time and MySQL Enterprise Thread Pool for increased server performance.

Along with faster data processing, MySQL Enterprise Edition High Availability has put an end to downtime, resulting in increased reliability of LocoTrack's real-time tracking of train velocities captured at train gates, intersections, and stations via the Global Positioning Systems (GPS).

Of critical importance to operations, the company reduced the time to back up 1TB of data from one full day, using Commvault - a plug-in to MySQL Community Edition's backup process - to between one and two hours with MySQL Enterprise Backup.

The railways organization also highlights the benefit of MySQL InnoDB Cluster, which provides all the benefits of high availability in an easy-to-use way with MySQL Shell. This allows KAI to scale the database workloads and achieve higher performance.

In 2021, KAI will add more database instances for its disaster recovery site and deploy InnoDB Cluster across all instances for increased data processing performance, resilience to failures, and the ability to perform scheduled maintenance without downtime.

What KAI considers a major advantage of moving to MySQL Enterprise Edition is the round-the-clock technical assistance and guidance of Oracle Premier Support. "The MySQL Support team is available to us 24x7, for best practices advice on database tuning and other measures to get the best speed and performance out of our systems," said Husni Rijal, Senior Database Administrator, PT Kereta Api Indonesia.

KAI's IT team were pleasantly surprised to find that MySQL Enterprise Edition provides a complete HA solution, including monitoring, backup, and security-the whole rolling stock of high availability. "MySQL is highly performant, the Enterprise Edition 8.0 is very powerful, and the security features are excellent. Furthermore, we now have improved backup, monitoring and support," Husni Rijal concluded.

Why MySQL Enterprise Edition?

With data being collected hourly from 200 locomotives and 500 railway stations, KAI needed maximum availability and performance of its databases, as well as first-class technical support. For these reasons, it upgraded from MySQL Community Edition to MySQL Enterprise Edition, starting with version 5.7 and moving to version 8.0 for enhanced clustering, security, monitoring, partitioning, and backup capabilities in 2020.

KAI will be increasing its deployment of MySQL Enterprise Edition across data centers and disaster recovery sites.

"With MySQL Enterprise Edition we have a far more powerful database, with greater security, performance, and reliability." Husni Rijal said.