Application Load Balancer Capacity Calculator

LCUs = max(new connections ÷ 25, active connections ÷ 3000, GB per hour ÷ 1, rule evaluations per second ÷ 1000). An Application Load Balancer measures four things every hour — new connections, active connections, processed bytes and rule evaluations — converts each into LCUs against its own allowance, and charges you for the largest of the four. Rule evaluations are requests per second multiplied by the rules processed beyond the ten that are free, which is why a long listener rule chain can quietly become the dimension you pay for even when bandwidth is modest.

Knowing which dimension drives the bill tells you what to optimise: connection reuse and keep-alive for the connection dimensions, compression for bytes, and consolidating listener rules for evaluations.

This calculator takes 6 inputs: New connections per second, Active connections in a one-minute window, Processed bandwidth, Requests per second, Rules processed per request, Price per LCU-hour. The pre-filled defaults are a realistic starting point — replace them with figures from your own environment for a result you can act on.

Because only the highest dimension counts. Short-lived connections without keep-alive can push the new-connection dimension to dozens of LCUs while bandwidth sits near zero, and you pay the connection figure.

Yes. The processed-bytes allowance is 1 GB per hour per LCU for instance, container and IP targets, but only 0.4 GB per hour when the target is a Lambda function, so the same traffic costs more LCUs. Confirm the current allowances and the LCU price for your region with AWS.