SNI ended the era of one IP address per HTTPS site. What replaced it needs judgement about multi-tenancy rather than TLS: which certificate answers a hostname you have never heard of, why forty customer domains should not share one certificate, and what you have quietly pooled by putting everyone on one address.
Virtual servers, pools and monitors map across almost mechanically. What stalls a BIG-IP migration is a folder of iRules nobody has catalogued — and the quiet discovery that several backends have been authenticating clients by source IP. A concept-by-concept map, an honest account of what does not translate, and a staged path that keeps rollback cheap.
Users losing their logins or baskets mid-session is usually a persistence problem. How to choose between source IP and cookie persistence, what each costs you, how to set the timeout from the application rather than a round number, and how to test the case that actually bites — a backend failing.
Your load balancer watches every connection to every service you run — which makes it the best vantage point in your infrastructure. The metrics worth watching, the patterns they reveal, and how to feed them into your monitoring platform over SNMP.
Where TLS ends determines what your load balancer can do. A decision guide to the three SSL modes — termination, passthrough, and re-encryption — and the feature, performance, and compliance trade-offs of each.
Traffic layers rarely fail loudly — they fall behind quietly. Five warning signs that the setup which got you here won’t get you further, and what the next step looks like for each.
Global Server Load Balancing steers users to the right data centre before a connection is ever made — using nothing more exotic than DNS. How it works, what it’s for, and how it fits alongside your local load balancers.
A load balancer that protects you from server failures but has no failover of its own is a liability. How active-passive HA with VRRP works, what actually happens during failover, and how to set up a pair.
Round Robin, Least Connections, weighted variants, and hash-based schedulers each suit different workloads. A practical guide to picking the algorithm that matches how your traffic actually behaves.
Free software is not free infrastructure. Where the real money goes in a DIY traffic layer — engineering hours, incident exposure, and key-person risk — and what consolidating it onto one platform changes.
