Uploaded July 2026 | Updated November 2026, 3 hours ago
When the application stops working, the first thing that we blame is the network. Whether its 2am on a Sunday or 11am Tuesday, network engineers are pulled in to answer the question "is the network down?" We throw on our hero suit in the phone box and start the timer on the only metric we seem to track: Mean Time To Innocence. In this talk I will show how at Proton, we leverage the same ideas highlighted in Google's Site Reliability Engineering concept to make networks and the applications they support more reliable by embracing failure. We will see how the Google concept applies to the network in design and operations, how observability is a must, and how to integrate all that into the applications on top (Spoiler alert, its not that hard).
#AutoCon5 #nre #networkreliabilityengineering #sre #sitereliabilityengineering
When the application stops working, the first thing that we blame is the network. Whether its 2am on a Sunday or 11am Tuesday, network engineers are pulled in to answer the question "is the network down?" We throw on our hero suit in the phone box and start the timer on the only metric we seem to track: Mean Time To Innocence. In this talk I will show how at Proton, we leverage the same ideas highlighted in Google's Site Reliability Engineering concept to make networks and the applications they support more reliable by embracing failure. We will see how the Google concept applies to the network in design and operations, how observability is a must, and how to integrate all that into the applications on top (Spoiler alert, its not that hard).
#AutoCon5 #nre #networkreliabilityengineering #sre #sitereliabilityengineering










