Re-architected a legacy vehicle booking platform on AWS Lambda, Aurora, DynamoDB, and CloudFront. The serverless approach eliminated infrastructure overhead, improved availability, and slashed operating costs.
Airtrans Limo and Tours runs one of the top 10 largest chauffeured fleets in the country and is the largest provider of luxury transportation in metropolitan Toronto. Since incorporation in 1998, the company has grown from 5 vehicles to 250 vehicles today.
Their on-demand vehicle booking platform was built on legacy infrastructure and development tools, with three mobile front-end clients—rider, driver, and admin—all dependent on it. As ride volume grew, the platform was experiencing system crashes, multiple errors including regular database failures, and slow data loading into client applications. The team needed to overhaul their backup and recovery procedures and improve Recovery Time Objectives and Recovery Point Objectives.
Airtrans wanted to implement best practices to efficiently process all booking requests and fleet and driver management data—without the constant operational firefighting.
When Cloudism was engaged to fix the stability and performance of the existing platform, we determined that fixing it would require more effort, time, and money than building a serverless solution from scratch. Cloudism developed a Proof of Concept using AWS Lambda in two weeks to demonstrate the benefits—faster development, easier operational management, automatic scaling, and reduced operational costs. We combined Amazon Aurora and DynamoDB to provide better business continuity and stronger Recovery Time and Recovery Point Objectives.
Pleased with how the team designed and developed the POC while keeping deliverables on track, Airtrans approved building the enterprise-level solution. Cloudism built it primarily using Lambda, API Gateway, DynamoDB, Aurora, Amazon S3, CloudFront, SNS, and Amazon Cognito User Pool.
/rest/*) to API Gateway and serves static content from an S3 bucket.After processing between 300,000 and 500,000 requests per month, fewer than 0.0002% of requests were dropped. Performance tests revealed significant improvements in latency, with data loading into client applications reduced to less than 3 seconds.
We are very pleased with how the Cloudism team designed and developed the application while keeping deliverables on track and cost under control. With AWS Lambda, Aurora, and SNS, we can respond to our riders and drivers more effectively, ultimately reducing response times and increasing customer satisfaction. Although our knowledge of AWS was very limited, Cloudism guided us through this cloud journey with their exceptional experience and professionalism.
Tell us about it. We'll help you scope a path forward grounded in real-world experience like this.