Back
Key Takeaway
Service speed innovation that reduced 15-second response delay to 0.02 seconds
Realized average service response speed of 20ms or less with a container environment reduced by 83% compared to the existing system, maximizing large-scale traffic processing performance and operational efficiency.
Airlines (B Company)
Client :Airlines (B Company)
Industry :Airlines / Transportation
Service Area :Migration & Modernization / Applications & DevOps
1. Overview (Project Background)
B Company pursued a project to «All-in Migration & Modernization» of customer-facing systems and internal systems that were previously operated in IDC environments to the cloud according to its cloud adoption strategy. Beyond simply moving infrastructure, the company aimed to secure system stability and scalability through application modernization and build an environment capable of flexibly responding even during promotional periods with large-scale traffic.
2. Challenge (Problem Definition)
B Company faced the following technical challenges and issues before and during the cloud transition.
Service Response Delay Issue
Intermittent 15-second response delays occurred in travel agency (OTA, Online Travel Agency) integration services, leading to degraded customer experience and persistent service quality issues.
Lack of Flexible Scalability
During traffic surges from promotional events, immediate server expansion was difficult, creating structural limitations in handling large-scale access.
Operational Inefficiency and Lack of Visibility
Rollback after deployment was difficult, version management was complex, and operational productivity was degraded as logs had to be checked by accessing individual servers.
3. Solution (Resolution Approach)
MegazoneCloud executed digital transformation across infrastructure, applications, and databases through a «Cloud Native Modernization» strategy that went beyond simple migration (Lift & Shift).
App Modernization and Resource Optimization
Converted applications to a container-based modern framework environment to streamline performance, and significantly optimized the number of key service containers from 24 to 4, maximizing resource efficiency.
DB Architecture Improvement and Zero-Downtime Migration
Separated customer-facing service and OTA (Online Travel Agency) databases to ensure stability, and achieved zero-downtime migration by leveraging AWS DMS CDC functionality for real-time data synchronization.
DevOps and Monitoring Environment Enhancement
Implemented deployment automation and rapid rollback system through ArgoCD-based GitOps pipeline, and secured operational visibility through integrated log monitoring via Grafana Loki.
Large-Scale Traffic Flexible Response System
Established ALB Pre-warming and NodeGroup Scale-out strategies to prepare for surging traffic during major promotions, ensuring service continuity with flexible expansion possible within minutes.
4. Result (Achievements)
Through successful cloud transition and architecture enhancement, the following business and technical achievements were realized.
99.8% Service Response Speed Reduction
Completely resolved the intermittent 15-second response delay issue that occurred in the existing travel agency (OTA) integration service and implemented a high-performance service maintaining average response speed of 0.02 seconds or less during normal operations.
83% Application Resource Optimization
Through application containerization and optimization, significantly reduced the number of containers for major services such as homepage and mobile from the existing 24 to 4, maximizing approximately 83% resource efficiency.
Enhanced Promotional Response Capability
Prepared a Scale-Up environment capable of expanding servers within minutes and extending database connection capacity to a maximum of 6,000 during large-scale promotional events, ensuring business continuity.
Improved Operational Productivity and Deployment Stability
Automated the existing manual deployment method based on open source to increase deployment accuracy and speed, and improved operational efficiency by reducing log inquiry and failure root cause analysis time through an integrated monitoring environment.







