Network redundancy exists where technologies and strategies are implemented within a network to preserve network stability by reducing or eliminating disturbances and downtime. It requires securing secondary network resources - including hardware and diverse network connections - over which data can be seamlessly rerouted when primary paths are unavailable for any reason. Network redundancy can be seen as an extra level of security, with focus on mitigating network outages and maintaining business continuity. Essentially, it is a form of contingency planning and is often part of an overall 'disaster recovery' plan. This is especially important for enterprises who rely on business-critical connectivity.
Where network redundancy is established with a parallel 'fail-over' system that handles data traffic during an outage, an additional level of security and reliability can be achieved by building in full geographical network diversity - particularly for mitigating service interruptions caused by catastrophic man-made or natural disasters. Here, the primary and secondary network service infrastructure - hardware, network cabling and the power resources are fully duplicated and physically separated.
Network redundancy requires thorough planning and preparation. This starts with the identification of the most prioritized segments, where interruptions have a critical impact on the network, and the most vulnerable segments where network failures are likely to cause service disruption.
It is then necessary to establish alternative network resources that can seamlessly carry network traffic in the event of disruption to the ordinary paths. The key components of a redundant network are:
Backup equipment that can carry traffic in the event primary hardware fails.
Alternative routes in the network that maintain connectivity when primary paths fail. This can be enhanced with diverse routing intelligence for even greater resilience.
Backup power is crucial for ensuring continuous operation during power outages. Uninterruptible Power Supply (UPS) systems or generators are a key component of the overall redundancy plan.
The main goal of our redundancy planning is to minimize customer service disruption by seamlessly rerouting data to alternative Points of Presence (PoPs) in the event of an outage. We achieve this by installing redundant equipment and links at every core PoP. By doing so, we enhance the overall resilience of our network.
Additionally, we offer a range of solutions across our global network that allow customers to select the best level of redundancy for their particular needs.
In most of the Arelion network, we own the underlying DWDM and fiber layers, ensuring the best possible capacity and route diversity. Our planning team manages IP, DWDM, infrastructure (space and power) and fiber holistically, and as such, network design and growth plans are readily shared across the team. On leased capacity sections, we work closely with carefully selected partners to ensure vendor diversity and that customers can turn up capacity whenever they need it.
Arelion’s network has continuously expanded over three decades, reaching new locations and countries with secure Points of Presence (PoPs). By focusing on customers, we meticulously design our PoPs to deliver the best possible network reliability.
Resiliency and redundancy are fundamental to our network design principles and by deploying diverse routes and PoPs, we minimize disruption and optimize data flow. We have contingencies for a multitude of physical and virtual threats, including vandalism and a range of natural disasters. We incorporate robust security measures including alarm technology, stringent access control and meticulously planned route diversity.
Our Internet solutions are designed so that traffic chooses the 'best and closest' connection choice, with inherent redundancy in the logical IP layer, insuring uninterrupted access to the Internet.
Ethernet connections provide both point-to-point (EVPL) and any-to-any (ELAN) high-bandwidth options via Ethernet over MPLS. These services leverage the inherent reliability and resilience of the network underlay. Ethernet, as a private network connection, ensures the availability of multiple routes for customer traffic, so in the event of a network failure, automatic traffic rerouting takes place. Additionally, with segment routing and flex algo low latency routes are consistently chosen for optimal latency.
Customers also have the flexibility to choose multiple access ports, along with redundant devices on-site which feature dual power supply configurations that can be tailored to specific site conditions.
Wavelength route diversity, often referred to as "optical diversity" or "fiber route diversity" involves the implementation of multiple physical paths, or routes, to send data over different wavelengths in multiple optical communication systems.
The above Arelion solution demonstrates how customer data centers in the UK and Germany are connected with a high-capacity dual-link 400GE. Here, traffic is distributed across both links during normal operation and can be carried on a single link in case the second one fails. A third link serves as a ‘belt-and-braces’ last resort in the unlikely event of complete dual-link failure. This is just one of many possible solutions where redundancy can be employed as a safety measure to minimize the risk of interruption as far as possible.
For more details about Arelion’s solutions, contact our sales and solution experts.
