Where Does Reliability Come From?

• Architecture (i.e. System Design): component redundancy as appropriate

  • What are common single points of failure? Sites, lines, routers, hubs, NICs, cables, servers.

  • Don't forget the mundane: AC power, HVAC, etc.

  
  
  

• Policies: what to do in case of component failure

• Practice: testing and maintaining the architecture and policies