Posts about Backend

At trivago we store a subset of our realtime metric data in InfluxDB and we are quite impressed by the load it can handle. Despite all the joy, we had to learn some lessons the hard way. It is pretty easy to overload the database or the web browser by executing queries that return too many datapoints.

At trivago we rely heavily on the ELK stack for our log processing. We stream our webserver access logs, error logs, performance benchmarks and all kind of diagnostic data into Kafka and process it from there into Elasticsearch using Logstash.

The advances and growth of our Selenium based automated testing infrastructure generated an unexpected number of test results to evaluate. We had to rethink our reporting systems. Combining the power of Selenium with Kibana’s graphing and filtering features totally changed our way of working.

Caching data is an essential part in many high-load scenarios. A local 1st-level cache can augment a shared 2nd-level cache like Redis and Memcached to further boost performance. An in-process cache involves no network overhead, so the cache speed is only limited by local resources like CPU, memory transfer speed and locking.

Learn how we managed to move fast and create a new Symfony application without breaking our old legacy session handling. We write to our legacy session (which is file based) from our new project which uses PDO as the session storage.

Here at trivago we write a huge number of log messages every day that need to be stored and monitored. To handle all these messages we created Gollum, a tool that enables us to conveniently send messages from multiple sources to different services.

At trivago we love hotels above everything else, but we also like metrics, we love to measure everything, compare, decide, improve and then rinse and repeat. In this blog entry we are going to describe our experience with InfluxDB, a time series database that we are using to store some real time metrics.