Predictable cost
XLB uses a flat negotiated license. Requests, connections, bytes, and traffic growth do not increase the software fee.
XLB vs AWS Network Load Balancer
AWS NLB makes Layer 4 load balancing convenient. At millions of requests per second, its hourly, connection, and traffic dimensions can become a major infrastructure bill. XLB delivers extreme-scale L4 performance on compute you control, with a flat license and no usage-based load-balancing fee.
Why teams choose XLB
Both products remove proxy tuning from the application team. The difference is the economic model. AWS operates a metered service for you. XLB gives your team a tuning-free product that runs efficiently on your own infrastructure.
XLB uses a flat negotiated license. Requests, connections, bytes, and traffic growth do not increase the software fee.
The XDP fast path routes packets in the Linux kernel without copying application payloads through a userspace proxy.
Point XLB to static backends or a Kubernetes Service. Backend discovery, updates, metrics, and autoscaling signals are built in.
| What matters | AWS Network Load Balancer | XLB |
|---|---|---|
| Pricing model | Load balancer hours plus the highest applicable NLCU usage dimension | Flat negotiated software license plus your underlying compute |
| Traffic growth | Connection and processed-byte growth can increase service cost | No request, connection, byte, or traffic-volume software fee |
| Infrastructure | AWS managed service | Kubernetes, bare metal, or virtual machines you control |
| Backend updates | AWS target groups and service integrations | Static configuration or automatic Ready Pod discovery from a Kubernetes Service |
| Autoscaling | Capacity scales inside the managed AWS service | Multiple XLB instances can scale from exported CPU and network metrics in Kubernetes |
| Monitoring | AWS service metrics and integrations | Built-in instance console plus premade Grafana and Datadog configurations through OpenTelemetry |
| Performance tuning | Handled by AWS | Completely tuning free |
| Commercial support | AWS support plan and service documentation | Commercial product with deployment assistance and ongoing support |
The cost difference
AWS bills NLB usage across hourly and capacity dimensions. That model is convenient, but a high sustained connection rate or traffic volume can turn the load balancer into a six-figure monthly line item. XLB replaces that variable service fee with a compact server footprint and flat commercial license.
This is practical high-scale guidance based on XLB’s smaller server footprint and underlying compute costs. Region, traffic shape, discounts, and hardware determine the exact result.
Kubernetes discovery keeps Ready Pods current automatically. CPU and network utilization metrics support horizontal scaling of the XLB tier.
XLB performs backend selection and packet rewriting in the Linux kernel. Application payloads do not travel through a userspace proxy process.
Common questions
At high scale, XLB typically runs at around 20% of equivalent managed network load balancer cost in practice. XLB has a flat negotiated software license and runs on compute you control. AWS NLB charges for load balancer hours and Network Load Balancer Capacity Units, with additional data transfer and public IPv4 charges where applicable. Your exact result depends on traffic shape, region, discounts, and server choice.
Yes, for supported IPv4 and TCP workloads. XLB can watch a Kubernetes Service, discover Ready Pods, update the backend pool automatically, and expose CPU and network metrics for autoscaling multiple XLB instances. A representative evaluation confirms the networking and availability design for your cluster.
No. XLB is commercially licensed for a flat negotiated fee. It is not metered by request, connection, byte, or traffic volume. You still pay for the servers and network infrastructure where XLB runs.
XLB is completely tuning free. You provide the listener and either a static backend list or Kubernetes Service reference. XLB handles backend updates and selects the fast-path configuration without application-specific proxy tuning.
Official product references
Competitor details on this page are based on the vendor’s own documentation.
Prove it with your workload
Bring your current topology, traffic profile, or cloud bill. We’ll scope a representative XLB evaluation.