Blazing-fast has become the standard for any app or search engine. Consumers get twitchy whenever there is even the slightest delay – reports show half of consumers would wait just 5 seconds and a third would leave the site for a competitor’s if it was too slow. But instant shouldn’t come at a cost – relevance and accuracy are just as important. This is especially true for flight search.
Amadeus, already a veteran in search technology, has delivered an innovative solution with Instant Search. This new technology unlocks the ability to search billions of travel options across multiple dimensions – cost, distance, location, date, route – in a matter of milliseconds – and book the most suitable flight.
It’s a simple idea, but complex to implement. Browsing billions of options across multiple criteria in real time requires serious performance. That is why Amadeus is increasingly counting on MongoDB as the database powering the latest generation of its feature-rich travel applications, now including Instant Search.
MongoDB gives Amadeus the flexibility to accelerate time to value and handle key data structures at immense scale, for the industry’s most demanding travel companies. Travel site KAYAK is already using Amadeus Instant Search technology to improve user experience for travellers and increase conversion rates.
In fact, as a central part of its digital initiatives, Amadeus has become one of the most advanced users of MongoDB, one of the world’s most popular modern database which is part of a new generation of non-relational technologies that don’t model data in the traditional way. Rather than using rows and columns to organise data MongoDB lets developers use documents which store information in a more natural way. Being engaged in such a range of complex and industry-leading projects has led the company to win MongoDB’s Innovation Award in 2015. Many best practices, and some features, have been created in concert with Amadeus staff.
On instant search for instance, proactive alerts generated from automatically generated monitoring data ensures any potential performance bottlenecks or issues are spotted well before they affect customers. To scale effectively, Amadeus has distributed the database across multiple shards, which means that the database deployment is broken into smaller more manageable parts. Using a WiredTiger storage engine, the team has also managed to compress storage by 80%, leading to significant cost reductions and performance improvements.
Aside from Instant Search, Amadeus is also working with MongoDB to build more than a dozen applications such as an accounting data warehouse that can power real-time analytics for airlines on their revenue and accounting practices.
In a world where milliseconds count, instant search gives travellers just what they need, when they need it. We are incredibly proud to work with Amadeus on such an expansive project and are excited to be part of building a personalised and connected travel future.