Load Balancer Distribution
Calculate how traffic is distributed across backend targets by a load balancer.
About Load Balancer Distribution
Simple round-robin load balancing divides total request rate evenly across all backends: requests_per_backend = total / backends. Weighted distribution instead allocates traffic proportionally to each backend's assigned weight (typically reflecting relative capacity, e.g. a backend twice as powerful gets a weight of 2 versus 1 for standard instances): backend_rps = total × (weight / sum_of_weights). Weighted round-robin is standard practice when backend instances have heterogeneous capacity, ensuring more capable instances handle proportionally more traffic instead of being artificially limited to an equal share.
Formulas
Equal (round-robin) distribution
requests_per_backend = total_requests / number_of_backends
Weighted distribution
backend_requests = total_requests × (weight / sum_of_weights)
Worked example
With the default values below, Load Balancer Distribution returns Requests per Backend (Equal Split) of 2,500.00. Change any input above and every figure here is recalculated in your browser.
Inputs used
- Total Requests per Secondminimum 1
- 10,000
- Number of Backendsaccepts 1–1,000
- 4
- Backend Weights (optional, comma-separated)
- 1, 1, 2, 2
Results
- Requests per Backend (Equal Split)
- 2,500.00 req/s
- Highest-Loaded Backend
- 3,333.33 req/s
- Lowest-Loaded Backend
- 1,666.67 req/s
- Sum of Weights
- 6
Step by step
- 1Values used
- 2Equal (round-robin) distribution
- 3Weighted distribution
- 4Requests per Backend (Equal Split)= 2,500.00 req/s
- 5Highest-Loaded Backend= 3,333.33 req/s
- 6Lowest-Loaded Backend= 1,666.67 req/s
- 7Sum of Weights= 6
Assumptions and limits
- The result is derived only from the 3 inputs listed above. Anything not among them is not part of the calculation.
- In particular, this calculator does not take inflation, tax, fees or charges or compounding frequency into account.
- Figures are rounded for display to 2 decimal places, so totals built from several rounded parts can differ slightly from the unrounded arithmetic.
How these calculators are built and tested is described on the Methodology page. If a figure looks wrong, please report it.
Frequently Asked Questions
When should I use weighted instead of simple round-robin?
Use weighted distribution whenever backends have different capacities — e.g. mixing larger and smaller instance types, or gradually shifting traffic during a blue-green deployment — so more capable (or more trusted) backends receive proportionally more traffic than smaller ones.
Does round-robin guarantee perfectly even real-world load?
Not necessarily — round-robin distributes requests evenly by count, but if request processing times vary (some requests are much heavier than others), backends can still end up unevenly loaded in terms of actual resource consumption despite receiving an equal request count.
How do health checks affect this distribution model?
This calculator assumes all backends are healthy and receiving traffic. In practice, load balancers continuously run health checks and remove unhealthy backends from rotation, which increases the effective load on the remaining healthy backends beyond what this static calculation shows.
What's the difference between this and least-connections load balancing?
Round-robin and weighted round-robin distribute based on a fixed ratio regardless of current backend state. Least-connections (and similar dynamic algorithms) instead route each new request to whichever backend currently has the fewest active connections, adapting in real time to actual load rather than a static split.