When you activate auto-scaling in Waves, AWS monitors you application and automatically adjusts capacity to maintain steady, predictable performance at the lowest possible cost.
To activate auto-scaling go to your environment, servers tab.
In the load balancer section make sure you have load balancer activated.
Activate auto-scaling by toggling the indicator.
Here you’ll notice you have 3 different options to scale by:
CPU: % of it’s usage.
Latency: Seconds of latency.
Number of request: number of requests per second.
The right selection will depend on your app, so you’ll have to test the different options with different threshold until it works fine for you.
An option that I recommend is setting the scale indicator to CPU with an upper activation at 50% and lower activation to 20%. This will add a server when the CPU is getting high and remove a server when is back low.