Practical guide
ALB pricing: turn hourly capacity into monthly LCU-hours
calculate Application Load Balancer LCU-hours. Includes a worked example, calculator scope, and practical questions.
Updated · Sources checked
What an LCU-hour means
Application Load Balancer charges combine running time and capacity usage. An LCU is determined by the highest applicable normalized traffic dimension in an hour, rather than by adding every dimension together. Relevant dimensions include connections, processed bytes, and rule evaluations; their definitions depend on the workload and target type.
The calculator expects LCU-hours already derived. It is a cost converter, so entering raw requests or GB in the LCU field produces an incorrect estimate even if the number looks plausible.
Worked example: add hours after selecting each maximum
Assume the applicable normalized dimension values in one hour are 0.4 for new connections, 0.7 for active connections, 1.2 for bytes, and 0.3 for rule evaluations. The hour contributes 1.2 LCU-hours, not 2.6.
For a hypothetical 30-day traffic pattern, suppose each day has 20 hours at one LCU and four hours at five LCUs. Daily usage is 20 × 1 + 4 × 5 = 40 LCU-hours. Monthly usage is 1,200 LCU-hours. A continuously running single ALB contributes 720 load-balancer hours during those 30 days.
The modeled subtotal is 720 × hourly rate + 1,200 × LCU-hour rate. Those usage totals can be entered directly. The example is constructed capacity usage, not a prediction from a particular requests-per-second value.
Why the hourly shape matters
Different dimensions can dominate at different times. Selecting a maximum from monthly average connection and byte values can hide those switches. Prefer hourly measurements or billing-derived LCU totals, and retain peak periods when building a new-service scenario.
The tool omits transfer and public IPv4 charges. Capacity reservations and features with separate charges also require their own review. It does not infer TLS, target-type, or rule behavior from a website’s visitor count.
For a useful comparison, vary one driver at a time. Compare before and after LCU-hours for a caching or connection-reuse change, then check whether the dominant dimension actually decreased.
Frequently asked questions
Should I add connection LCUs and byte LCUs? No. Apply the documented maximum rule within each billing interval.
Does twice the request count mean twice the ALB bill? Not necessarily. The fixed hourly charge and the dimension that dominates capacity both affect the result.
Calculate your scenario
Use the application load balancer cost calculator. Keep the displayed region, pricing date, billing units, and exclusions alongside your result. The arithmetic examples above illustrate usage or explicitly hypothetical rates; they are not AWS quotes.
Source
Billing structure checked against official AWS documentation on September 6, 2026.