When Things Break
Host: Ulf Wiger
Scalability and fault-tolerance requirements have now become mainstream problems, and service disruption can easily prove fatal to your business. Things break, and quite often, they break at the most inopportune time - for example, nearly half the downtime in the US telephony network is attributed to overload.
In this track we will explore strategies for making systems "failure-tolerant", as in: tolerant to external failures as well as internal ones, even in the face of massive scale and ever-changing requirements.
