Building Secure and Reliable Systems at Scale
summary
Argues that security and reliability are inseparable and gives concrete Google-era fixes: replace fragile VPNs with SSL-based BeyondCorp (pioneered 2011) to stop lateral movement; break monoliths into containerized microservices (example uses Kubernetes) so failures and breaches are contained and can be redeployed in minutes; remove direct root access with a proxy requiring security keys and Zero Touch Prod two‑party approvals. Cites an SRE study: 0.01% error rate per engineer implies ~75% chance of a major outage annually at scale.
AI-written summary · about
chapters (22)
- 0:00Introduction
- 0:23What is reliability
- 2:23Supply chain attacks
- 4:29What does this mean
- 5:34Enterprise Networks
- 10:46Zero Trust Networking
- 12:40SSL
- 14:30Application Development
- 16:49Problems with Open Source
- 19:34Microservices
- 20:48Monolithic Systems
- 22:21Modern Architecture
- 26:44Mobile Phone OS
- 28:17Managing Infrastructure
- 29:55Google Study
- 30:44Zero Touch Prod
- 31:48Proxy
- 34:07Security
- 35:00Supply Chain
- 36:20Starting Small
- 37:54Book Club
- 38:28Additional Resources
related talks
43mSecurity By Design
23mScalable security: how to win friends and not burn...
31mRadical Ideas Enterprises Can Learn From The Cloud
48mSecurity today: change your habits and make it safer and efficient
24mSecurity by default
36mHow to build continuous security and embed it in your product
33mSecurity and SRE: Natural Force Multipliers
42mEnterprise Architecture Beyond the Perimeter