Load Balancer

What it does#

The Load Balancer stands for a reverse proxy such as NGINX that spreads requests across your App Servers.

It takes every request kind and sends it on. It adds a little latency and has a throughput limit, which each size step doubles. It spans every zone, so a zone outage does not take it down.

By default it splits traffic by each target's capacity, so a large App Server gets a larger share. Round robin is also on offer. The Load balancing algorithms Research adds least connections and power of two choices. Two choices handles more traffic with a little imbalance.

It drops a dead host at its next health check. The Health checks Research lets you change the interval and turn on deep checks. A Load Balancer can also route by request kind: set which kinds each of its links carries.

When to place it#

Place it when you need a second App Server, or when hosts behind DNS have mixed sizes. Resize it when its card shows it near its throughput limit.

How it fails#

• Overloaded, it drops requests before they reach any App Server.

• A long health check interval keeps routing to a dead host during an Instance failure.

• A deep check takes out every App Server whose Database is down or overloaded. If that is all of them, the Load Balancer has nowhere to send traffic, and a slow Database becomes a full outage.

Tips#

• Use least connections when App Servers have mixed sizes or runtimes.

• Route slow kinds, such as exports, to their own App Servers.