How do companies use AWS to increase their agility and resilience? Case Study - Expedia Group
As an International company with numerous partners also operating under the Expedia brand, Expedia Group acts with the EWE(Expedia Worldwide Engineering) organization to support its own websites and those of its partners. There were problems with a suggestion service that was supposed to improve the customer experience by recognizing the advance input of website visitors and speeding up the application. To do this, the so-called ESS(Expedia Suggest Service) uses customer location-based algorithms and aggregated customer shopping and booking data to display suggestions when a customer begins typing.
However, the ESS did not provide the result Expedia wanted. According to the company's metrics, an error page is the primary reason potential customers leave the site. EWE's engineering team found that ESS needed to operate in locations that were physically close to customers to provide a fast and responsive service with minimal network latency. However, because the data center used was solely focused on the Chandler, Arizona area, this resulted in excessive latency and an unsatisfactory customer experience on the sites.
To solve the problem, Expedia Group partnered with AWS (Amazon Web Service) as it was the only provider with the required global infrastructure for customers in the Asia-Pacific region. Thus, Expedia Group was able to solve the problem through AWS's International Infrastructure and hence proximity to the customer.
Using the innovative foundation of AWS, the ESS service was developed and deployed within three months. Expedia Group first launched ESS in the Asia-Pacific region and then quickly replicated the service in the US and also the EU regions. On the AWS-based foundation, average network latency was reduced from 700 milliseconds to less than 50 milliseconds, resulting in fast website load times, improving the visitor experience.
Expedia Group uses AWS to develop applications faster, to process large amounts of data, to scale, and to troubleshoot issues quickly. When creating a standard deployment model with AWS, development teams can quickly create infrastructure for new initiatives. Key applications run in multiple Availability Zones in different regions to ensure data is always available and enable disaster recovery.
Today, through collaboration and the concept of serverless cloud computing, Expedia Group no longer needs to provide in-house servers that are only used when fully utilized, but are still a drain on the budget in terms of acquisition and management. With AWS, they don't have to pull capacity or provision extra, they can add new capacity and only pay for what they actually use.
Magesh Chandramouli, ESS Principal Architect:
"When I was a developer, you didn't want to invest in architecture without knowing what the application would look like. I had to plan ahead and create a feasibility study for everyone's reference. With AWS, I'm not bound by throughput constraints or CPU capacity. When I think of AWS, the word freedom immediately comes to mind."
Over the next few years, Expedia Group plans to migrate 80 percent of its mission-critical applications from its on-premises data centers to the AWS cloud. By using AWS, Expedia Group has become more resilient to failures. Expedia Group developers have been able to innovate faster, saving the company millions of dollars.