A Travel Tech Company Undergoes a Seamless Scale and Performance Manageability for Their Tech Platform

The client is one of Indonesia's largest and most comprehensive online travel companies, catering to a wide range of travel and lifestyle needs. As a new-age Online Travel Agent (OTA) and lifestyle app, the platform boasts a global user base of 17 million and continues to grow by delivering seamless and innovative solutions for modern travelers.

The Problem Statement

The client faced critical challenges in ensuring consistent performance and reliability across all verticals during high-traffic sales events. They needed a robust pre-production environment and efficient handling of events during peak periods. Additionally, issues with their GCP account, including memory rightsizing and compatibility, highlighted the urgent need for a more organized and scalable infrastructure.

Challenges

Ensuring consistent performance across all verticals during peak traffic sales events in October and December.

Supporting end-to-end handling and management of events during sales periods.

Addressing multiple issues with the GCP account, including memory rightsizing and compatibility challenges.

Establishing a well-organized and robust pre-production environment.

Resolving cost optimization and implementing enhanced security measures.

Solutions

To address the client’s challenges and enhance platform efficiency, scalability, and security, the following solutions were implemented:

Streamlined GCP account management for better resource organization and utilization.

Set up Eagle Eye P0 Alerting to improve system visibility and incident response.

Simplified the integration and management of open-source tools such as ELK Stack, Kafka, Argo CD, Jenkins, Rundeck, Consul, MongoDB, and HAProxy.

Enhanced security by implementing IAM policies, IP whitelisting with HAProxy, and migrating code from shared servers to GitOps for better change tracking.

Optimized pre-production environments with automation in performance testing, code versioning, and an improved deployment process.

Resolved GCP challenges, including insufficient quotas and load-balancing issues.

Enabled cost optimization by creating automated jobs to notify about running instances during non-working hours.

Established a production-like performance testing environment by creating database backups from production and restoring them for reliability testing.

Additionally, migrated services from VMs to Kubernetes for improved scalability and efficiency.