I’m a research scientist working at WhatsApp Infrastructure, experienced in cluster scalability, transport protocols, container management, server efficiency, and control planes.
I’m also very interested in distributed protocols and algorithms, and I like to solve hard technical problems.
In this talk, we will cover multiple areas in WhatsApp’s infrastructure related to the lifecycle of a request flowing between our Erlang services. It includes how to overcome the limitation and scale our cluster to more than 40 thousand nodes using Erlang Distribution, the service discovery and routing mechanism, the job dispatching and queueing model, as well as the bottlenecks and overload protections.
We welcome you to come and learn on how to build Erlang infrastructure at scale.
AUDIENCE: Developers who are interested in large scale Erlang clusters