How it works

Reachability becomes another measured path.

Nexus Atlas does not bolt a separate tunnel migrator onto its data plane. Relay and direct connectivity enter the same link scheduler, so upgrading or falling back does not rebuild the secure Atlas tunnel.

STUN adoptedTURN replacedICE replaced
Connection sequence

Five stages, from unknown network to resilient bond.

The relay provides immediate rendezvous and reachability. Direct-path discovery happens after communication already works.

01

Map

Fetch and verify the signed relay inventory.

02

Connect

Open outbound authenticated sessions to the selected relay set.

03

Classify

Run STUN from the same socket that carries Atlas data.

04

Rendezvous

Exchange candidates and attempt the same ordered paths together.

05

Schedule

Measure every working path and apply the chosen traffic policy.

01 · Relay-first

Connectivity does not wait for a successful hole punch.

Both nodes initiate outbound sessions to an authenticated relay. CGNAT normally permits that traffic, so rendezvous and tunneled communication can begin without an inbound rule or public address.

outbound onlypinned relay keywarm NAT binding
02 · Direct upgrade

The probes that test the endpoint are also the hole punch.

Nodes exchange server-reflexive and local candidates through their existing relay session. They try the same candidate order together. When a response returns, the direct path becomes alive with a real measured RTT.

no reconnectno path nominationfailed punch costs nothing
03 · Continuous choice

A direct path is preferred because of measurements—not because it is called direct.

Direct and relay paths are ordinary Atlas scheduler inputs. If cellular routing makes the relay faster, or if the direct mapping disappears, policy can move traffic without invoking traversal logic again.

RTTjitterlosscapacity
Path behaviour

Direct and relayed states are deliberately uneventful.

A change in path availability changes scheduler inputs. Applications keep using the same Atlas interface and addresses.

Network conditionDirect pathRelay pathAtlas behaviour
Endpoint-independent mappingLikely availableWarmDirect usually wins on measured delivery time.
Endpoint-dependent CGNATUnavailableActiveRelay continues carrying the tunnel.
Mobile address changesRe-evaluatedAuthenticated roamingNew valid sources are adopted per path.
Direct degradationScore worsensAlready measuredTraffic shifts according to policy.
Primary relay failureUnaffected if liveSecondary electedConfigured relay redundancy remains available.
Standards relationship

Familiar building blocks, optimized for Atlas endpoints.

Both ends are Nexus Atlas, so interoperability machinery intended for unrelated clients can be replaced by simpler product-native components.

S

STUN · adopted

RFC-compatible address discovery and mapping classification. Atlas relays provide the observation points.

T

TURN · replaced

An Atlas-native forwarder carries opaque encrypted frames without allocation and channel-binding complexity.

I

ICE · replaced

Relay rendezvous discovers candidates; the existing Atlas probes and scheduler decide how every live path is used.

Get early access

Bring us the network you cannot control.

Apply as a pilot partner if you have a deployment ready to test, or start evaluating Nexus Atlas Traversal as a future customer. Tell us enough to understand the endpoints, carriers, regions, and operational constraints involved.

01

Pilot partner

Run an early deployment with direct engineering contact and help shape priorities through real-world feedback.

02

Customer evaluation

Research architecture, security, deployment options, and commercial fit before committing to a pilot.

Already enrolled? Open the console
Early-access applicationUsually answered by the product team

Your application is sent to the Nexus Atlas product team. You can also email office@nexusatlas.io.