Enhancing the Performance of Serverless Computing Frameworks for High-Demand Web Applications

With serverless computing, developers can build applications through code while experts handle the supporting infrastructure. Cloud Function services from AWS, Google, and Azure help businesses grow while staying affordable but heavy traffic makes these systems slow to activate and poorly scalable. The performance limits at these usage levels are hard to handle in e-commerce stores and video streaming platforms that need fast network response. This research applies optimization methods to improve serverless platforms when handling web applications with high traffic. The research implements scale prediction, warm start activation, and resource distribution technology to resolve key performance limitations in serverless platforms. Our test system worked with all three cloud providers: AWS Lambda, Google Cloud Functions, and Azure Functions. We used Apache JMeter and Locust to reproduce web application demands during testing. The project measured response time and scalability alongside throughput and cost metrics before and after serverless framework optimizations. The results show significant improvements: Our analysis revealed that response times decreased by 52% throughput soared by 127% and the system became 21% more cost efficient. The results show how serverless technology deals with large user loads better at a lower cost. Our research teaches useful ways to improve serverless architecture performance for actual applications. Our study helps improve serverless technology by fixing its main performance problems to make it more dependable and scalable. Future research will build better traffic prediction methods and combine architectural designs to create better application performance results worldwide.