Networking

Keep Services Available Under Any Load

Distribute incoming traffic across instances automatically. SSL termination, configurable health checks, and session persistence are available from day one.

Layer 4 & Layer 7

TCP load balancing for raw throughput, and HTTP/S routing for host or path-based traffic rules in the same service.

Automatic Health Checks

Unhealthy backend instances are removed from rotation and brought back automatically once they recover.

SSL / TLS Termination

Offload TLS handshakes to the load balancer and keep backend traffic on the private network.

What You Get

Layer 4 and Layer 7 balancing
SSL/TLS termination
Health checks
Session persistence
Backend pools and members
WebSocket support

Best For

High-traffic websitesAPI gatewaysMicroservicesScalable applications

Put multiple instances behind one entry point, with health checks and TLS termination.

Technical Specifications

Protocols
HTTP, HTTPS, TCP, UDP
Algorithms
Round Robin, Least Connections, IP Hash
TLS support
TLS 1.2 and TLS 1.3
Health check interval
Configurable - minimum 5 s
Session persistence
Cookie-based or source IP
Backend model
Listeners, pools, and members

How It Works

1

Create a load balancer

Choose the protocol, port, and balancing algorithm.

2

Register backend instances

Add the instance IPs that should receive forwarded requests and group them into a backend pool.

3

Configure health checks

Set the check path, interval, and healthy or unhealthy thresholds for automatic failover.

4

Point your DNS and go live

Create an A record pointing to the load balancer floating IP and put traffic in front of the pool.

Common Use Cases

High-Traffic Web Applications

Handle traffic spikes without leaving a single instance exposed as the bottleneck.

E-commerce peak trafficMarketing campaign launchesFlash sale events

API Gateways

Route API traffic across several instances with health checks and controlled entry points.

REST API clustersGraphQL endpoint poolsWebhook receiver farms

Rolling Deployments

Drain one backend, deploy a new version, then re-register it without exposing users to a full outage.

Blue/green deploymentsRolling instance upgradesCanary traffic splits

Need the Right Starting Point?

Tell us what you run today and we will suggest a realistic starting point, not an inflated setup.