Please log in to watch this conference skillscast.
Aggregating reponses used to involve only a single database join; now information is retrieved from multiple services by making several API calls, introducing complexity and latency. Caching becomes much more important to ensure our new aggregation-based service can deliver results in a performant, responsive manner. But where should we create this cache? How long should we cache for? Should our services be aware of it? What software and architecture should we use when creating our caching platform?. We will talk about our approach, and our use of NGINX, Memcached and Couchbase to boost the performance of our services.
YOU MAY ALSO LIKE:
The Stockholm MicroSyndrome: Missing my database inner join
Tom LIllington
Thomas Lillington is a Developer at Net-a-Porter. He joined the company as a fresh-faced graduate, having an opportunity to work in many different languages and on many projects both large and small. A relatively recent convert to Scala he is now finding his way in a functional, SOA world.
Matthew Green
Matt came through the graduate scheme at Net-A-Porter having been fresh out of university, and now works for the NAP Tech team. As a full stack developer he is currently working on a new service-based platform for the user-facing website, predominantly in Node.js, but also with a bit of Scala and Java.