After isolating production, office and device networks: Who is dual-OLT redundancy designed to protect?

At 2 a.m. in the workshop, injection molding machines were running automatically. A sudden jitter on the backbone network halted the entire production line for three minutes. For precision assembly lines, this window is long enough to scrap a batch of work-in-progress parts. Many factory managers assume that separating production, office and device networks via VLANs is all that is needed. Isolation does block lateral threats such as an infected office PC spreading malware to PLCs over the network. However, isolation cannot prevent links themselves from failing.

Three-network isolation addresses “cross-network intrusion”; dual-OLT and dual-core redundancy addresses “service outage”. These are two distinct requirements. The former builds a security boundary, while the latter safeguards business continuity. Without either one, production lines may shut down unexpectedly.

1. Three-network isolation: Block unauthorized cross-access first

Factory networks host three types of traffic with vastly different characteristics.

• The production network connects PLCs, SCADA systems and CNC machine tools. It is sensitive to jitter, and even millisecond-level command fluctuations can spoil batches of materials.
• The office network serves hundreds of employee laptops and mobile devices, with video streaming and downloads as common traffic patterns.
• The device network (also known as OT or IoT network) carries cameras, access control units and environmental sensors. It features heterogeneous hardware, legacy protocols and widespread weak passwords, making it a common attack entry point.

If these traffic types share the same broadcast domain, incidents are inevitable. A compromised office endpoint may spread laterally into the production network and lock PLCs; a hijacked camera can be used to reach core switches.

AINOPOL’s integrated connectivity-security solution separates the three networks through VLAN isolation at Layer 2. Production traffic is segregated from office traffic, entertainment traffic cannot consume dedicated production bandwidth, and visitors are assigned independent user groups. Isolation forms the primary baseline for separating production, office and guest access.

This three-network setup runs on AINOPOL’s all-optical foundation. The two-layer flat architecture cuts computer room space and energy consumption by over 70%. POF optical-electrical composite cables deliver combined power and signal transmission in one cable. This all-in-one cable resists strong electromagnetic interference and supports transmission over 800 meters without repeaters. Instead of stitching together hardware from three different vendors, network segmentation is configured once within a unified all-optical framework.

2. Dual OLT / dual core: Insurance against link outages

Once isolation is in place and cross-access blocked, links can still fail. Buried fiber may be cut, optical modules age, or main control boards reboot. Any single point of failure can take the production network offline.

A dual-OLT (e.g. MA8500X) + dual-core (e.g. CS8500E) + dual firewall cluster deployment provides backup paths.

How does this redundancy work? Two OLTs operate under Type B/C dual-homing protection. If one goes offline, the other completes failover within 50ms. Core switches adopt dual power supplies and dual main control hot standby; production continues normally even if one power module or main board fails. Further upstream, dual firewalls in an active-standby cluster ensure seamless takeover if one perimeter security device fails. The 50ms threshold is critical. Production control cannot tolerate momentary disconnection. Failover taking several seconds will generate defective products, while a 50ms switchover is transparent to the production line. The whole network eliminates single points of failure from access layer to core and perimeter, supporting 7×24 non-stop operation and avoiding mass line stoppage caused by isolated hardware faults.

3. Redundancy targets critical workloads, not a one-size-fits-all full deployment

Redundancy resources should not be spread evenly across the whole network. A half-hour outage of administrative OA systems causes minimal disruption, but a fraction-of-a-second loss of real-time control commands for injection molding machines creates scrap. Dual-homing protection, dual power supplies and dual main control units should be prioritized for the production network and critical equipment links. This “outage insurance” is deployed where downtime carries the highest cost.

Office and guest networks can tolerate second-level jitter, so their redundancy tier can be reduced as appropriate. This balances reliability and investment, without requiring dual-homing for every branch link.

This architecture also requires differentiated service grading. AINOPOL’s application-aware intelligent bandwidth scheduler prioritizes production control, video conferencing and ERP traffic for low-latency forwarding. It raises overall bandwidth utilization from under 40% to over 80%. Critical services are protected first at the link layer, then at the bandwidth layer; redundancy and traffic scheduling are two sides of the same design.

These isolation and redundancy capabilities are not assembled from disjointed boxes requiring separate integration. The AINOPOL Dream Gateway integrates routing, AC controller, firewall, IPPBX and OLT in a single appliance. Multiple buildings and workshops are centrally managed from headquarters. The network topology is visualized clearly, faults can be located with one click, around 80% of faults can be resolved remotely, and O&M is supported via mobile APP.

Built on all-optical infrastructure, powered by audio & video engine, and protected by robust security capabilities, AINOPOL delivers an integrated “connectivity & security” solution. VLAN isolation, dual-homing protection and perimeter firewalls are natively enabled within one unified platform. Network administrators no longer need to switch between management consoles from multiple suppliers. Expansion is achieved by adding devices on demand, and policy templates can be reused for new workshops. Isolation and redundancy are designed as two complementary elements from the initial planning stage.

FAQ

Q: Should VLAN isolation or dual-OLT redundancy be implemented first?
A: Isolation is recommended first to draw clear security boundaries. Once isolation is stable, dual-homing and power redundancy can follow. The two can be rolled out in phases and do not require simultaneous deployment. Prioritizing isolation mitigates the highest lateral movement risk first, while redundancy addresses business continuity afterwards. The sequence is not mandatory, but starting with isolation is more practical.

Q: Does the dual-core setup require two CS8500E switches?
A: The model selection depends on site assessment. The core requirement for dual-core deployment is dual power supplies, dual main control boards and elimination of single points of failure. The exact configuration is determined by access scale, port density and future evolution requirements and documented in the project proposal. Whether to deploy dual cores and how many units are needed depends on the production line’s tolerance for interruption.

Q: After deploying dual OLTs, do regular switchover validation tests still need to be performed?
A: The value of dual-homing protection lies in reliable failover during real failures. Whether to run validation tests and the test methodology fall under operational procedures. It is recommended to align with the solution provider and include validation clauses in the delivery agreement.