Building High-Availability Services with Linux



Free yourself from worrying about whether that one machine with a critical service is going to go down by using open-source tools to implement automated failover. Expand capacity and improve reliability for your services by spreading the load over many inexpensive servers.


I’ll provide an overview of what “high availability” means (and doesn’t mean), system and network architectures for implementing high availability, and Linux-based tools for implementing failover and load-balancing (Linux-HA/heartbeat, ldirectord, LVS/IPVS). I’ve used these tools to implement high-availibility service implementations for HTTP, SMTP, POP, and IMAP.

Speaking experience