by Antoine Grondin (@aybabtme) on Thursday, 25 February 2016
View session in schedule
- Full talk
- Technical level
teach the benefits of designing software in a crash only fashion, with examples from large scale deployments at DigitalOcean. how to design for graceful handling of failures.
I will talk about what crash only software is, why you want it and how to implement it. we have real examples of crash only software saving our sanity in multiple large scale deployments at DigitalOcean.
Antoine is a distributed system engineer at DigitalOcean working on an upcoming storage product. Previously he worked on a distributed caching layer for disk images, and before working at DO, on a distributed load testing service on the Performance team at Shopify.