Modern software has thousands, perhaps millions, of moving parts. Your users could be all over the world, accessing your system from devices you’ve never even heard of. Even the simplest systems often have dozens of components: controllers, models, databases. With cloud hosting and “software as a service”, it’s trivial to spin up and plug in specialist services like media transcoder pipelines, NoSQL data stores, API gateways – and as systems grow more complex, it gets harder and harder to understand what they’re actually doing.
Metrics and monitoring is a vital part of site reliability engineering. Visualising the components in your system, and the traffic flowing between those components, can be the key to identifying performance and capacity problems. If you’re affected by a site outage or a suspected data breach, accurate metrics is essential to understand the severity of the incident and plan your response.
I’ve worked with systems from ELK (ElasticSearch, LogStash, Kibana) and NewRelic, through bespoke monitoring systems like PRTG and AWS CloudWatch, to mass-market web analytics tools like Google Analytics and HotJar. Whether you want to understand your website traffic, identify performance bottlenecks in your system or get your budget forecasts under control, get in touch and find out how metrics and monitoring can help.