Dedicated capacity
The wavelength carries your traffic and nothing else. There is no oversubscription ratio to ask about, because there is nothing to oversubscribe.
Capacity that belongs to one customer.
A wave is a dedicated optical wavelength running between two facilities. It sits below Ethernet switching, so there is no shared equipment deciding whose frames go first and no other customer able to consume capacity you are paying for. The port rate is available to you, continuously, whatever anybody else is doing.
That is the difference from a switched Layer 2 circuit, and it is the reason to buy one. For most interconnection, a point-to-point Layer 2 circuit is the better answer — quicker to turn up and cheaper to run. A wave earns its cost when the traffic is heavy and sustained, when the application cannot tolerate contention, or when you want the path to yourself for reasons of your own.
Because the service is transparent, what you put on the port is what comes out the other end. Your framing, your MTU, your Layer 3. We do not participate in your network.
At a glance
| Service | Dedicated optical wavelength |
| Speeds | 10G / 100G |
| Hand-off | Ethernet client port at each end |
| Protection | Unprotected or protected |
| Transparency | Full — your framing and MTU |
| Coverage | Nationwide, select facilities |
| Contract | One, with Network Edge |
| Support | One NOC, one ticket |
Speeds and protection
10G and 100G, protected or not.
The hand-off is a standard Ethernet client port at each end, so a wave lands in your equipment the way any other optical circuit does. No special optics on your side beyond what the port rate calls for.
Unprotected or protected
Unprotected is the default and suits customers who build their own diversity. Protected is available where the underlying route supports it, and two unprotected waves on separate paths is often the better answer when you want diversity you control.
Protocol transparent
The service sits below Ethernet switching. Your framing, your MTU, your Layer 3 — unchanged end to end. Nothing in the path inspects or rewrites your traffic.
Reach
42 markets, coast to coast.
Where a wave can be delivered, and the national corridors that join those markets.
Markets where a wave can be delivered, joined by the national long-haul corridors between them. This is a reach map, not a route map — it shows where a circuit can go, not which fiber a given circuit will ride. The path a particular wave takes is designed and confirmed when it is quoted.
SeattlePortlandSacramentoSan JoseLos AngelesSan DiegoLas VegasPhoenixSalt Lake CityBoiseDenverAlbuquerqueEl PasoSan AntonioAustinHoustonDallasOklahoma CityOmahaMinneapolisKansas CitySt. LouisMemphisNew OrleansNashvilleChicagoIndianapolisDetroitColumbusClevelandAtlantaJacksonvilleTampaMiamiCharlotteRaleighPittsburghBuffaloAshburnPhiladelphiaNew YorkBoston
Availability is per building, not per city — see the buildings we serve in each market.
Nationwide, at select facilities.
Waves are available nationwide at select facilities. Reach depends on the buildings at both ends, so send us the two addresses and we will confirm what is deliverable and on what timescale.
Waves are delivered over partner and carrier optical networks that Network Edge arranges, manages and supports. We are straightforward about that, because it is what determines where a circuit can go: reach is a question of which buildings the underlying networks are in, not of a coverage map we would like to draw.
What you get from us is the part that is usually the hard part. One party designs the circuit, orders it — across more than one underlying provider where the route calls for it — manages the turn-up, and answers the phone afterwards. You hold a single contract and open a single ticket.
What we need to quote
- The two facility addresses, with suite or cage if you have them
- Port rate: 10G / 100G
- Protected or unprotected
- Whether diversity from an existing circuit matters
- Target turn-up date
- Term you want priced
Use cases
When a wave is the right tool.
Data-center interconnect
Two facilities behaving as one, at full port rate, without the traffic between them competing with anything else.
Replication and backup
Bulk transfer that has to complete inside a window, where a contended path turns a nightly job into a morning problem.
Reaching a market you are not in
Extend into a facility where you have no equipment, and take colocation, transit or exchange access there once you arrive.
Carrier and backhaul capacity
Move aggregate traffic between your own points of presence without buying a share of somebody else’s switched network.
Cloud on-ramps
Dedicated capacity into the facility where your cloud interconnect lands, sized to what the workload actually moves.
Disaster recovery
A path that behaves the same on the worst day as on an average one, which is the only day the DR site matters.
Wave or Layer 2? Usually Layer 2.
We would rather sell you the right circuit than the larger one. A point-to-point Layer 2 circuit covers most interconnection, turns up faster and costs less. It is the default recommendation and it is what most customers should buy.
Come to a wave when a specific thing is true: the traffic is heavy and sustained rather than bursty, contention is unacceptable for the application, or you need the path itself to be yours. If you are not sure which side of that line you are on, describe the traffic and we will tell you honestly — including when the answer is the cheaper product.
Side by side
| Layer 2 circuit | Wave | |
|---|---|---|
| Path | Switched Ethernet | Dedicated wavelength |
| Capacity | Committed rate | Full port rate, always |
| Contention | Shared equipment | None |
| Turn-up | Faster | Longer |
| Cost | Lower | Higher |
| Best for | Most interconnection | Heavy, constant traffic |
Questions
Frequently asked
What is a wave?
A dedicated optical wavelength between two facilities. The capacity is yours alone for the life of the circuit — it is not a share of a larger pipe, and no other customer’s traffic rides on it. Because the service sits below Ethernet switching, whatever you put on the port comes out the far end unchanged.
How is a wave different from a point-to-point Layer 2 circuit?
A Layer 2 circuit is Ethernet carried across switching equipment, which is the right answer for most interconnection and is usually quicker and cheaper to turn up. A wave is a dedicated lambda: no shared switching in the path, no contention for capacity, and the full port rate available at all times. Choose a wave when the traffic is heavy and constant, or when you need the path to itself.
What speeds do you offer?
10G / 100G. The hand-off is a standard Ethernet client port at each end, so it lands in your equipment the same way any other optical circuit would.
Where can you deliver one?
Waves are available nationwide at select facilities. Reach depends on the buildings at both ends, so send us the two addresses and we will confirm what is deliverable and on what timescale.
Can I get a protected circuit?
Yes. Unprotected is the default and the usual choice for customers who build their own diversity. Protected waves are available where the underlying route supports it, and we can also quote two unprotected waves on separate paths where you want the diversity under your own control rather than the carrier’s.
Who do I call when it breaks?
Network Edge. The circuit may cross more than one underlying optical network, but you hold one contract with us and open one ticket with our NOC. Chasing the underlying provider is our job, not yours.
Can a wave carry my IP transit?
Yes. A common design is a wave into a facility where AS7201 is present, with transit turned up over it — so your router does not have to sit in the same building as the hand-off.