Urgent.News

What's breaking now, across thousands of outlets.

Tech

The LCU Trap: Why Your Load Balancer Bill Has Nothing To Do With Bandwidth

` Most engineers assume load balancer costs scale with throughput: "We push 2 Gbps, so our bill should reflect 2 Gbps." That intuition is wrong. AWS doesn't bill ALB or NLB by throughput. They bill by LCU (Load Balancer Capacity Units) —and you pay for the single largest dimension , not the sum. The Rule: Pay for the Max, Not the Sum Four dimensions are evaluated every hour: New connections/flows…

Many engineers mistakenly believe that load balancer costs scale linearly with throughput. They expect to pay directly based on the amount of data being transferred. However, this is not how Amazon's ALB (Application Load Balancer) or NLB (Network Load Balancer) are charged. Instead, AWS bills by Load Balancer Capacity Units (LCU).

The rate is determined by the single largest dimension among four factors evaluated hourly: new connections/flows per second, active connections/flows, bandwidth (in Gbps), and rule evaluations per second (only for ALB, beyond 10 free rules).

The key rule governing billing is that customers are charged for the maximum of these dimensions, not the sum. You pay for the highest demand, regardless of how other factors compare. This means optimizing the other metrics has no impact on your bill.

To illustrate the difference, consider two scenarios:

1. A high-connection workload:

- ALB requires 2,000 LCU for 50,000 new connections per second, costing around $11,712 per month.

- NLB requires only 63 LCU for the same volume, costing approximately $308 per month – a 32 times better efficiency.

2. Rule evaluation costs:

- Every rule beyond the first 10 incurs additional LCU cost calculated as (rules - 10) * rps / 1,000.

- With 5,000 requests per second and 50 path-based routing rules, rule evaluations alone require 200 LCU, amounting to $1,168 per month, despite negligible bandwidth usage.

Customers can use an interactive calculator on jslet.com to input their specific traffic profile and see which dimension is binding them – the single factor determining their actual cost.

Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Read the original at dev.to →

More in Tech

Proxmox: Wann LXC-Container und wann KVM-VMs die bessere Wahl sind

Proxmox: Wann LXC-Container und wann KVM-VMs die bessere Wahl sind Einleitung: Die Entscheidung, die jedes Proxmox-Team plagt Wer Proxmox einsetzt, steht irgendwann vor der Gretchenfrage: Soll ich…

  • LXC offers direct device access, improving I/O operations over KVM VMs
  • KVM provides better security boundaries and isolation than LXC

Four Failures I Kept Repeating on AWS Amplify — Full env Replacement, Green Build but 500 at Runtime, Silent Outages, Domain Migration

From April to July 2026, I ran several Next.js products on AWS Amplify Hosting (AWS's managed service for hosting and deploying web apps).

  • Four distinct failure categories identified in AWS Amplify usage
  • Failure 1: update-app command replaces entire env map, wiping existing variables
  • Failure 2: Build passing does not guarantee runtime functionality, especially in SSR

More from Thursday 27 August →