All complex computer systems eventually break, despite all of the heavy-handed, bureaucratic change-management processes we throw at them. But some systems are clearly more fragile than others: it all depends on how well they cope with stress. In this O’Reilly report, Dave Zwieback explains how the DevOps methodology can help make your system antifragile.
Systems are fragile when organizations are unprepared to handle changing conditions. As generalists adept at several roles, DevOps practitioners adjust easily to the fast pace of change. Instead of attempting to constrain volatility, DevOps embraces disorder, randomness, and impermanence to make systems even better.
This concise report covers:
- Why Etsy, Netflix, and other antifragile companies constantly introduce volatility to test and upgrade their systems
- How DevOps removes the schism between developers and operations, enlisting developers to deploy as well build
- How to use continual experimentation and minor failures to make critical adjustments—and discover breakthroughs
- How an overreliance on measurement and automation can make systems fragile
- Why sharing increases trust, collaboration, and tribal knowledge