SK Planet Builds Real-Time Data Analytics Environment Across the Company by Adopting Trino

SK Planet adopted open-source Trino as its company-wide data analytics engine to shorten data lead time and strengthen real-time analytics on its big data platform DIC (Data Integration Cluster). The existing Hadoop-based environment had operational data loaded on a D-1 (previous day) basis, causing degraded data freshness and wasted resources when moving large volumes of data.

Trino allows access to various data sources simply by connecting to them, without data migration or loading, making it optimized for joins across heterogeneous DBs and real-time queries. SK Planet used Trino Gateway to eliminate a single point of failure (SPOF) and implement cluster redundancy, and integrated Apache Ranger with LDAP to precisely control data access permissions for company-wide members.

As a result of the adoption, faster performance than existing Hive was confirmed for processing small-volume data under 10 million Rows. This shortened the reporting cycle for marketing targeting and service analytics from daily to hourly, and by integrating with the in-house data catalog system MeDic, users can now utilize real-time data with only SQL syntax changes.