Yedekli Network

 

By using Ethernet networks to manage our industrial processes and mission critical infrastructure systems, we are able to increase productivity and efficiency. But the more we rely on the communication systems, the greater the demands are on reliable solutions. We have reached a point where these networks play such an important role in our world that failures have become unacceptable. The result of a system that does not work can be expensive, inconvenient, dangerous and damaging to the environment.

Westermo delivers resilience through WeOS, our software that operates on a range of our most robust hardware platforms. Highly reliable Ethernet switches running resilient network protocols dramatically reduce the risk of network failure.
Mission Critical infrastructure networks need resilience. They provide fresh drinking water to our homes and workplaces, 24/7 electricity, waste water systems, regular and reliable rail services, traffic control, global marine transportation of goods and more.

Westermo delivers resilience through WeOS, our software that operates on a range of our most robust hardware platforms. Highly reliable Ethernet switches running resilient network protocols dramatically reduce the risk of network failure.

FRNT Ring Coupling

 

 

On a solar plant the panels are in constant and controlled motion to ensure that they are always facing the optimum direction for power generation. During daylight hours every second earns money for the operator of the plant. A network failure will massively impact on the ability of the network to operate at 100% efficiency so resilience is critical.
If a failure does occur in a ring network then the plant will continue generating for the rest of the day until maintenance can be carried out overnight or at the next scheduled shut down.   It is during this period that the plant would be vulnerable to a second failure which is why a multiple resilient network is needed.
Network operation with multiple network media failures
The need

Mission critical infrastructure networks and industrial applications must operate for 100% of the required time for the systems they connect. Sometimes this can mean 24/7 or just during the operation for which they are critical.  Stopped manufacturing lines, interrupted infrastructure services, and other system failures are unacceptable as they inevitably result in costs for the operator.

If a single network failure occurs it is often not possible to carry out immediate maintenance. An engineer may not be available or the operator may not risk stopping a system that is currently working to repair a broken cable.

The processes running on mission critical networks often employ protocols that cannot tolerate packet losses of more than a few milliseconds before registering errors. Network errors will sometimes force an expensive  process shut down.

Networking consideration

If all data on a ring network was directed to known devices then the address learning capability of Ethernet switches would allow rings of basic switches to be used. Unfortunately many protocols generate broadcast traffic that floods to all parts of the network to find its destination. If normal switches are configured in a ring this data will circulate forever in the network, multiplying until eventually the network overloads.

Special protocols exist like FRNT that ensure that these broadcast storms never occur in a ring network.  When a failure occurs this protocol must rapidly reconfigure the network to ensure the process can continue.

For ultra-resilience networks must sometimes tolerate more than one failure. One method of building a resilient network is to use routers that learn paths through networks and rebuild these as the topology adjusts. Unfortunately these protocols like OSPF, RIP and BCP tend to have relatively long reconfiguration times as they were designed more for wide area networks. The automation type data transport protocol  (Modbus, Profinet, EthernetIP) often employed in these networks were originally designed for local networks and tend to flag faults if the recovery time of the network is too slow.

The solution

FRNT is a layer-2 network ring protocol that ensuring reconfiguration time less than 20ms in large networks.  FRNT Ring Coupling enables you to connect two FRNT rings, a sub-ring and a super-ring, with multiple layer-2 uplinks. This Ring Coupling functionality ensures that one of the uplinks is always active, while the other is in backup state, thereby providing redundancy and loop-free connectivity.  It is also possible to set priority on the uplinks to ensure the fastest path when different medias are used for the uplinks.

The ring coupling protocol reacts to link loss as well as an echo response from the interconnected switches. The uplinks topology can also be extended to connect rings in a ladder or tree structure.

With FRNT and FRNT Ring coupling protocol it is possible to design a network with multiple rings based on layer-2 technology. This then ensures a resilient network with fast reconfiguration time to all nodes even though multiple network failures occurs.

Multilink Dual< Homing

 

 

this fresh water application remote I/O´s have been used with only one Ethernet interface. Security and resilience are essential for this type of application. All installed equipment needs to have connection to both of the redundant rings, which WeOS provides. A major benefit of using a WeOS product is that with the ever increasing worry of cyber-attack the built in firewall means that a single device can solve two network issues in a simple reliable package.

Multiple connections to a single device
The need

When one network device, for instance an I/O block or industrial controller, is seen to be critical to a system it makes sense to consider providing dual connections to separate parts of a network. Some industrial equipment has two network interface cards, but in many cases this option is not available. Also with this solution the device has to support a protocol that decides on which network connection to use which can add complexity to the i/o function.

Networking consideration

In any Ethernet network it is important not to accidently create rings. If this happens then network broadcast packets can end up circulating the network endlessly resulting in the ‘broadcast storm’ phenomenon. To provide the most reliable solution it is important that the primary and secondary links connect to separate parts of the backbone network. This kind of connection is defined as dual homing.

The solution

WeOS offers a feature called Multilink Dual Homing that allows more than just a dual connection to the network. It is actually possible to configure as many as eight of these links back to the main network. With this function network solutions can be created offering multiple levels of resilience.

Obviously the switch that connects to the equipment can be seen as a single point of failure, much like the i/o device. It is therefore critical that this switch is designed with the highest level of reliability to ensure maximum levels of network resilience. The best measure of reliability is the MTBF of the product. The Westermo Lynx and Redfox range of products offer industry leading figures exceeding 500,000 hours.

FRNT Ring Coupling and Multilink Dual Homing

Safety is the overriding design criterion for the systems in a modern road tunnel. Ventilation is vital as is the management of the traffic entering the tunnel. In the event of an incident it is important to be able to prevent more people entering the tunnel. When this kind of safety is at stake the fault tolerance of the network is a mission critical consideration. By the use of Multilink dual homing together with FRNT ring coupling protocol a resilient rapid recovery network is possible.
Complex resilient network design with fast recovery time
The need

Some industrial networks are so critical that they must be resilient to both media and equipment failure. If there is a failure there may be no opportunity to carry out network maintenance for some time. For this reason therefore it is desirable to have a network that can tolerate multiple failures. The traditional approach to this is to use routing protocols, however these protocols can be slow to reconfigure the network  and also the network hardware tends to have a higher cost.

Networking consideration

Protocols such as OSPF and RIP have been designed for huge wide area networks. The protocols allow network hardware to be connected with multiple backup paths that are selected if the primary paths fail. As these protocols are designed for wide area networks they often take several seconds to reroute data in the event of a topology change. These router technologies are defined as a layer 3 function. For faster network reconfiguration layer 2 switch protocols need to be used to control the network topology.

The solution

A combination of Westermo’s Multilink Dual Homing and FRNT Ring Coupling provide a rapid recovery resilient network solution. Different media can also be incorporated to make use not just of normal copper and fibre network cables, but also twisted pair connections using SHDSL technology.

Benzer ürün ve yazılar

Arama