Case Study: KAIST achieves seamless high-concurrency enrollment and real-time performance with Altibase

KAIST adopts Altibase to resolve the strains of massive concurrent logon attempts while providing a solid performance

Korea Advanced Institute of Science and Technology (KAIST) was struggling with its online course enrollment system: its legacy DBMS could not handle massive concurrent logons during peak enrollment, causing widespread disconnections, downtime and degraded service as online courses and sessions increased. To resolve this, KAIST adopted Altibase, deploying its hybrid in-memory DBMS with replication features.

Altibase implemented a hybrid-mode, in-memory solution with active-active replication (deployed on a SUN FIRE 15k), providing high availability and seamless access to current and historical account data. As a result, KAIST now runs enrollment periods without interruption, supports over 1,000 concurrent sessions versus 300–400 previously, and increased throughput from 800,000 periodic transactions to 1.2 million transactions processed in real time.

KAIST