Richard Warburton
TOPIC: Fantastic Performance and where to find it
This example driven talk will guide you through concrete approaches to using continuous profilers in a production environment.
Remember the old days - when your software stack was simple to understand and requirements were less demanding? With modern systems distributed microservices talk to each other all the time over the network. Cloud deployments mean a huge variance in the type of hardware that you deploy to. Rapid iteration and deployment cycles result in rapidly changing workloads that invalidate performance tests before they’re written. Globalised, 24/7 uptimes mean that functional or performance outages aren’t acceptable at any point in time.
These changes have helped businesses get their product out to market much more rapidly but they often make it harder for developers to find and fix performance problems or to improve the efficiency of their systems. From attending this talk you’ll understand:
* Common profiling visualisations and reports
* How to use profilers to solve performance problems
* The benefits of continuous, exploratory, profiling
* How to improve scalability, have happier customers and reduce infrastructure costs