Service and license fees
Managed NLB charges grow with usage. Commercial proxy editions or enterprise support can add per-server costs as the fleet grows. XLB uses a flat negotiated license with no per-byte or per-flow fee.
Operating cost / extreme scale
Lower the cost of every request.
XLB is built for companies handling millions of requests per second. Move more traffic with fewer servers, remove traffic-based load-balancing fees, and reduce the engineering work required to keep the tier running.
Millions of RPSNo traffic feesFewer servers
The full cost
The load balancer bill has three parts.
At modest traffic levels, the difference can be easy to ignore. At millions of requests per second, service fees, server count, and operator time all become material infrastructure costs.
Server footprint
HAProxy and NGINX can require a larger proxy fleet at extreme scale. XLB is designed to move more traffic per load-balancer server.
Operational overhead
Fleet sizing, proxy tuning, backend automation, and control-plane upkeep consume engineering time. XLB removes most of that work.
Cost model
Three approaches. Very different bills.
Managed NLB shifts the work to a cloud provider but meters usage. HAProxy and NGINX avoid that traffic meter, but add compute, operational work, and potentially commercial licensing or support. XLB is designed to minimize the total.
| Cost driver | Managed NLB | HAProxy / NGINX | XLB |
|---|---|---|---|
| Service pricing | Usage grows with traffic and connections | Open source, or per-server commercial license and support | Flat license, no traffic-based fee |
| Load-balancer compute | Included in the service charge | Dedicated proxy fleet | Higher throughput per server |
| Performance tuning | Handled by the provider | Owned by your team | No application-specific tuning |
| Backend updates | Cloud integration | Proxy config or separate automation | Automatic from Kubernetes |
| Control-plane upkeep | Handled by the provider | Architecture dependent | No separate control plane |
High-scale cost guidance
~20%typical cost versus equivalent managed NLBBased on the smaller server footprint and underlying compute costs. Actual results vary by workload and infrastructure.Where the savings come from.
- 01
Fewer machines
XLB routes packets without copying application data through a userspace proxy, increasing throughput per server.
- 02
No traffic meter
Run XLB on infrastructure you control without a load-balancing charge attached to every byte or flow.
- 03
Less engineering time
Remove repeated tuning work, backend configuration maintenance, and a separate control plane from the operating model.
Cost questions
Validate the savings against your workload.
What does the 20% cost guidance mean?
At high scale, XLB typically runs at around 20% of equivalent managed NLB cost in practice. The estimate comes from the smaller server footprint and underlying compute costs. Traffic shape, hardware, region, pricing, and existing discounts determine the exact result.
What should we bring to a cost evaluation?
Bring managed load-balancer usage and pricing, or the server count and instance types in your HAProxy or NGINX fleet. Request rate, connection rate, average packet size, and peak traffic help scope a representative test.
Does lower cost require giving up Kubernetes automation?
No. XLB watches the Kubernetes Service selector, uses Ready Pods, and keeps the backend pool current as the application scales or rolls.
How is XLB software priced?
XLB is commercially licensed for a flat negotiated fee. The license is not metered by requests, connections, bytes, or traffic volume. Deployment assistance and ongoing support are defined in the commercial agreement.
Run the numbers
What is your load balancer costing you?
Bring your cloud bill or proxy topology. We’ll scope an evaluation around your real traffic and infrastructure.