One of the major challenges faced by teams working on high growth product is of performance. Systems that are built for a given scale of users often fail to deliver the necessary throughput when run with orders of magnitude of load more than what they are built for. Software teams have historically resorted to a myriad set of ways in scaling performance.
In this talk, I shed some light on how Slack scales its performance to its millions of global users. I cover the various aspects of Slack's performance scale, going in from the people aspect of it - where the focus would be on describing how several backend, frontend and SRE teams work in unison to solve scalability woes, to the implementation aspect of it - where the highlight would be on giving a technical picture of how it is done in a short example case study.