Disaster Recovery & Data Replication
-
Your Replication Window
The length of the route sets the replication window. On-net Wavelengths and Private Optical run a known, repeatable latency between your sites, and you see the measured figure for your route before you commit.
-
Production and Replica Apart
Physically diverse on-net routes reduce the chance that a single fiber event takes down production and its replica at the same time. One team owns and engineers the routes you run on.
-
Capacity for the Full Copy
Replication is bursty: steady change data plus periodic bulk syncs. Data Center Connect carries the bulk volume; dedicated Wavelengths carry the latency-sensitive paths. Capacity climbs as your data volumes do, up to 400 Gbps at the client handoff.
Your route sets the recovery floor
Disaster recovery depends on a copy of your data that sits far enough away to survive a local failure and current enough to recover from. Those requirements pull against each other, and the network between the sites decides the tradeoff.
A bank under a zero-data-loss mandate and a hospital replicating imaging between data centers hit the same wall: a synchronous write cannot be acknowledged faster than the round trip between the sites. How tight a recovery point you can target depends on that round trip.
When that network is leased and stitched together from several carriers, round-trip times drift between sites and the separation you were promised can share conduit you have no way to check.
One cut can take production and replication down together.
How Lightpath approaches it
Owned routes, dense where your sites are
- Lightpath owns, builds and operates more than 12,100 route miles, with the densest footprint across the Northeast and Mid-Atlantic: the New York metropolitan area, Long Island, New Jersey, Southern Connecticut, Boston, Eastern Pennsylvania and Ashburn. The network also serves Miami and is expanding into Columbus, Phoenix and Atlanta. That footprint reaches 190+ on-net data centers, where most production and recovery sites sit, and 17,500+ service locations. Where a recovery site falls outside the footprint, Lightpath delivers over Type 2 and coordinates the carrier so you keep one point of contact.
Latency that holds, because the path does
- Replication latency tracks the physical length of the route and how consistently that route performs. On dedicated Wavelengths and Private Optical, the path is engineered and owned end to end on-net, so the latency between two on-net sites is a measured, repeatable number Lightpath can put in front of your team. Latency is not quoted on off-net or Type 2 segments, because Lightpath does not own them.
Capacity at the right layer, with diverse routes underneath
- We hand off dedicated wavelengths up to 400 Gbps at the client interface, on a network engineered for 800 Gbps and above on the line side. Per fiber pair and conduit, the ceiling climbs past that as your volumes grow. Underneath, physically diverse on-net routes reduce single points of failure on the paths you depend on.
Engineers who own your routes
- When you are on our network, you reach the engineers who design and operate it directly. No offshore tiers, no ticket queues. With financial backing from Optimum, we keep investing in route density across the markets we serve.
Replicating to a cloud recovery target
Lightpath Cloud Connect delivers private paths to a cloud recovery target:
- AWS Direct Connect
- Microsoft Azure ExpressRoute & Gov’t ExpressRoute
- Google Cloud Interconnect
- Oracle FastConnect
- Equinix Fabric
Lightpath is the transport that carries replication traffic to the provider’s on-ramp; it does not replace the cloud provider’s storage or recovery product.
On the facility side, Lightpath is present in Equinix, Digital Realty and Markley Group across Boston, Ashburn, Chicago, Miami and Secaucus.
You get one Lightpath contract and one escalation path for the segment Lightpath delivers.
Case Studies
Built on these products
| Category | Product | Bandwidth | When to use it | |
|---|---|---|---|---|
| Connectivity | Wavelengths | 10G to 400G (client interface) | Dedicated Layer 1 transport between production and recovery sites where latency must be low and repeatable | |
| Connectivity | Private Optical | 1Gbps to 800 Gbps | Wave-based private network for replication across three or more sites | |
| Connectivity | Dark Fiber | Unlimited (customer managed) | When you want to own and operate your own optics and run your own replication wavelengths end to end | |
| Industry Apps | Data Center Connect | 20 Mbps to 800 Gbps | High-capacity interconnect for bulk replication and backup between facilities | |
| Industry Apps | Cloud Connect | Hosted 50 Mbps to 10 Gbps; Dedicated 1/10/100 Gbps | Private on-ramps to AWS, Azure, Google, Oracle, Equinix Fabric | |
| Managed Services | Optical Encryption | 10 Gbps and above | Layer 1 in-flight encryption for sensitive datasets | |
| Managed Services | Advanced Proactive Monitoring | N/a | 24/7 network health monitoring |
Don’t take the latency and diversity claims on faith. Verify them.
We will walk your team through the actual fiber topology for your routes, the measured latency between your named sites and the escalation model that names who owns your circuit at each stage. The route review is where the real numbers live, under a mutual nondisclosure agreement.
Request a route review
Get a quote
The fiber between your production and recovery sites should not be the part of your DR plan you cannot see or measure. Start with the route and the latency. Then talk pricing.
Call 1-877-544-4872 or complete the form.