For over 10 years we've developed ExchangeRate-API's infrastructure to deliver uptime you can rely on.
ExchangeRate-API has been carefully designed to deliver currency data every single time you send a request.
Day in, day out, 365 days a year.
Our infrastructure has been battle tested for over a decade and we've put all this experience into our stack and hosting choices.
In 2010 the first versions of our system were not very reliable and we had a number of outages...
The consequence was clear - many customers left after each downtime!
And as developers ourselves we understood exactly why, there are few things as frustrating as integrating with an API only to experience patchy connectivity and timed out responses.
So we set out on a mission to build an unbelievably reliable currency converter API and today things are very different!
Our product is hosted by top tier providers, has as few moving parts as possible and is carefully monitored in many different ways.
We care as much about our uptime as you do!
Amazon Web Services is one of the largest network and server providers in the world. Using AWS means we get access to best-in-class systems and exceptional uptime on our underlying infrastructure.
We also make use of the Availability Zone system AWS provides so that our system can run at the same time in multiple data centres that are physically and logically separate for high levels of redundancy.
ExchangeRate-API is delivered over the Cloudflare network. This mainly means exceptional network availability, but also protection from DoS attacks and access to latency reducing network features.
We have some early customers who have been using our API for nearly the entire decade we've run ExchangeRate-API. Our strong uptime record (99.98% uptime in 2019) is one of the reasons they've stayed!
You can only manage what you measure so we're always working to improve our monitoring & stats collection. We use Pingdom to run our API status page so you can see the live API status and uptime history.
We have curated our technology stack to include Open Source software that can really be depended on. There's nothing too new - we prefer battle tested classics like nginx, memcached, cron(8), etc...