Andrey Sibiryov
Andrey Sibiryov
Uber
Senior Infrastructure Engineer
Andrey Sibiryov is an SRE in Uber, New York. He is mostly focusing on monitoring & observability, performance, and infrastructure abuse issues. He was also the author & lead developer of the Cocaine Cloud platform. Current open-source projects include:
  • Kernel load-balancer project Gorb. Based on IPVS, the same technology that powers Docker Swarm Routing Mesh.
  • Tesson – NUMA-aware application sharding tool.
Previously, Andrey led the Cloud Technologies department at Yandex and worked on the [Helios](https://github.com/spotify/helios) CI/CD platform for Docker at Spotify.
Go Runtime & Wishful Thinking

Our industry has forgotten that software we write runs on hardware, no matter how many frameworks, containers and clouds we use in between, and, unfortunately, this abstraction gap is only growing. This talk about modern computers, wishful thinking in software engineering and performance is heavily influenced by our experience & observations running Go in production under extreme load. It encourages to be mindful about hardware and demonstrates why it’s important.