Canada's flagship carrier is the country’s largest domestic and international airline, serving more than 200 airports on 6 continents. They were facing challenges in terms of efficiency, capacity planning and performance optimization. HCLTech’s efforts helped the client see significant improvements in system efficiency, reducing costs and improving user experiences along the way.
The Challenge
Outdated tech stack with lack of integration capabilities
Our client faced several challenges across their technology stack, including:
Web application
- The website was slow, with the homepage taking nearly 30 seconds to load
- Users were facing poor experiences with web images loading
Mobile app
- User complaints regarding higher battery usage when the app was being used
- High mobile app crashes observed in the production and a poor rating in the app store
Middleware systems (Mulesoft and IBM MQ)
- New technology with IBM Messaging Queue (MQ) and Mulesoft
- Undefined and non-functional requirements related to capacity and resiliency
- High volumes of data feed processing at the data hub, which required sizing of channels and adoption of the compression algorithm
Digital notification platform
- Limited exposure to cloud transformation
- Loosely coupled micro-service architecture involving multiple components that required tuning and optimization
Contact center
- Migrating the IVR system from legacy to AWS connect platform, as it involved capacity planning, performance optimization and improving call quality
- Migration of the agent desktops to AWS AppStream
Azure AI
- Lack of observability
- Concerns in volume testing and capacity planning
- Synthetic data generation difficulties
The Objective
Improve system reliability and eliminate bottlenecks
The primary objectives were to improve web application performance, enhance mobile app responsiveness, optimize middleware systems and enhance various other aspects of their technology stack. They were looking to improve the user experience and address performance bottlenecks and reliability issues, as well.
The Solution
A holistic approach to address breadth of challenges
To achieve the outlined objectives, a comprehensive solution was implemented with a plan to target and improve all areas.
Web application
- Engineering analysis for the major webpages waterfall
- Recommended use of web accelerators for image and content optimization
- Optimization of the order of page loading and implementation of service workers
- Upgradation to H2 protocol to promote push strategies for faster page loads
- Optimization of caching strategy
- Conducted Selenium-based load tests to capture UX metrics
Mobile app
- Used Xcode and Android studio tools to conduct mobile app profiling and identify performance issues at the code level
- Recommended optimization techniques to reduce battery usage
- Enhanced app responsiveness through code and resource optimizations
- Optimized GPU usage for smoother performance
Middleware systems (Mulesoft and IBM MQ)
- Conducted thorough capacity planning exercises
- Implemented resiliency features such as redundancy and failover mechanisms
- Optimized channel sizing and implemented message compression
Digital notification platform
- Integrated complex event streaming systems involving Apache Kafka and Flink
- Implemented circuit breakers, autoscaling and throttling mechanisms
- Deployed tools for monitoring, tracking and measuring system observability
Contact center
- Optimized IVR for faster and more efficient call routing
- Conducted VDI capacity planning for improved performance
Azure AI
- Enhanced observability by creating dashboards with alerts
- Successfully completed volume testing
- Addressed capacity planning concerns
- Implemented synthetic data generation for testing
The Impact
Measurable improvements in nearly every KPI
This initiative not only improved system efficiency and reliability but also significantly enhanced the user experience across our client's diverse digital platforms. It showcased the value of performance optimization and capacity planning in ensuring the seamless operation of complex digital ecosystems.
The results included:
- Webpage performance: Improved the webpage loading time from 30 seconds to 5 seconds
- Middleware optimization: Achieved a 20 TPS (transactions per second) increase with IBM MQ compression and a 50% cost reduction with service virtualization
- Digital notification platform: Optimized complex event streaming, improved data availability, enhanced observability and implemented circuit breakers, autoscaling and throttling mechanisms
- Contact center: Realized a 20% improvement in call quality and an 18% reduction in call drop rates
- Azure AI: Attained reductions in processing and data generation times of 10% and 25%, respectively
