CIO Influence
Edge Computing Guest Authors

We Need to Think Differently – Your Hybrid Connectivity Isn’t Really Hybrid

We Need to Think Differently - Your Hybrid Connectivity Isn’t Really Hybrid

With advancements in technology, and in particular that which carries data across our ether, hybrid connectivity has become increasingly popular as a means to optimizing communications. Promising a seamless integration of multiple network technologies, it aims to provide enhanced reliability, flexibility, and performance. However, upon closer examination, what many believe to be hybrid connectivity, isn’t truly hybrid.

First and foremost, we need to think differently and take a close look at the ‘classic networking approach’, and how failover, whilst offering a degree of resilience and redundancy, isn’t truly hybrid. Despite its apparent benefits, hybrid does not truly embody the essence of hybrid connectivity. Failover refers to the ability of a system to switch to a backup connection when the primary connection fails. While failover mechanisms do indeed provide a level of redundancy, they do not fundamentally transform the nature of the connectivity architecture. As a consequence, the degree of resilience is inherently constrained.

CIO Insights: The AI-security Paradox: A Human Story of Security and Privacy

In a truly hybrid – or ‘heterogeneous’ – network, multiple network technologies seamlessly work together, actively sharing the load and resources, combining, and binding together a variety of bearers from cellular and LTE to satellite and WiFi into a single ‘pipe’. In this way, it can deliver a faster and, more importantly, more reliable service.

In fact, a truly hybrid platform should go a stage further than that, accommodating and configuring itself for a range of other variables, depending upon each bearer’s performance and other environmental conditions affecting it at any one moment in time in order to optimize performance and reduce costs – but I’ll touch on that later.

When optimized in this way, a hybrid system is capable of working through degradation and failure to ensure that single TCP connections are maintained and accelerated, regardless of the availability and performance of the underlying networks. This ability to optimize connectivity, even in the most challenging conditions, will deliver a truly consistent and uninterrupted user experience. In critical operations, where connectivity can be the difference between life and death, hybrid and resilience take on a whole new meaning.

Failover, on the other hand, relies primarily on a backup connection to take over in case of an outage, without actively utilizing the strengths and capabilities of both primary and secondary connections simultaneously.  It is the concept of failover, I would suggest, which challenges the traditional orthodoxy of what hybrid connectivity really means.

The Increased Reliance Upon Resilience and Redundancy

The need to be able to ‘connect’ within any given environment, regardless of whether terrestrial infrastructure is available or has been compromised is now more significant than ever. Live streaming of video, the transfer of data and voice calls are crucial within countless sectors.

Defense will always require multiple ways to communicate with fallback options. Emergency services need real-time access to ongoing incidents, whether that be drone video footage above a forest fire in a remote or rural area where cellular infrastructure isn’t an option. The ambulance service is reliant on good connectivity to access patient data or the services of a clinician while on the move or at the kerbside. Telehealth triage has moved at a rapid rate of knots and none of this tech will work without hybrid connectivity and the robust bonding of all and any available networks.

Likewise, broadcasters are ever more reliant upon critical live streaming from anywhere on the planet; the marine industry meanwhile is utilizing the power of hybrid connectivity as yachts and commercial vessels navigate the seas, and even utilities require machine-to-machine (M2M), and IoT capability to control and monitor operations on windfarms, oil rigs and installations that are traditionally ‘out of reach’.

With the relay of live video no longer the preserve of news organizations, but a mainstay of multiple sectors and consumers, the biggest challenges are no longer around relaying footage from A to B, but with the network and available bandwidth. Now, the key to delivering the highest quality video is bonding available networks to increase bandwidth, seamlessly correcting errors and optimizing the link.

Even the cars we drive and the appliances we use daily are harnessing variations of satellite, WiFi and cellular communications, and with low earth orbit (LEO) satellites becoming more and more accessible, the applications and opportunities are endless.

Factors such as latency, packet loss, and network congestion can hinder the integration of networks, resulting in sub-optimal user experiences. As a result, the envisioned performance improvements of hybrid connectivity become overshadowed by the complexities and compromises it introduces in the classic approach – hence a need to think differently.

Security and Compliance versus cost and QoS

Integrating different network technologies from diverse sources introduces additional security challenges and consideration across various network endpoints and connection points.

The lack of a unified security framework can result in vulnerabilities and inconsistencies that make the overall network infrastructure more susceptible to cyber threats. Moreover, ensuring compliance with industry regulations and data protection standards becomes increasingly challenging in a fragmented network environment which incorporates public networks and the Internet.

When it comes to being efficient with cost-management, a key consideration is that with true hybrid, heterogeneous connectivity, different parameters can be configured to allow for cost efficiencies to be made. For example, with the transmission of voice calls, parameters can be set to allow for the most cost-effective bearer to be used if it’s good enough – i.e., cellular can take preference over satellite if it’s ‘stable enough’, thus reducing the costs of always using satellite.

This also applies to the handling of Quality of Service (QoS), to ensure the performance of critical applications where there’s limited network capacity, rapid variations in bandwidth and latency across each network.  While security and compliance remain a network engineer’s priority in many organizations, it should never be compromised by cost. In a true hybrid ecosystem, this compromise need never be a concern because factors such as priority, efficiency and cost have formed part of the underlying design and mirrored in the configuration settings.

Why does all this matter?

The old adage says that if you do what you’ve always done, you’ll get what you always got. But that’s not the case in networking. If you do what you’ve always done in a networking environment, where the explosion of tech, amount of data being transferred and the sheer weight of traffic vying for space on our heavily congested networks, you will in fact get considerably less than you always got.

So, we need to think differently. Many hybrid systems are not the least bit hybrid, they simply provide a backup where the switchover is often measured in minutes. While this might be adequate for some, it doesn’t meet the evolving requirements of industries that rely upon always-on, fail-safe connectivity, as cited earlier.

To achieve full hybrid connectivity, and as part of a European Space Agency (EAS) contract, we developed RazorLink, an industry-first Software Defined Networking (SDN) solution that can seamlessly and dynamically bond any number of bearers, from satellite, cellular, Point to Point radio, Wi-Fi and terrestrial services in line with user-defined objectives and prevailing conditions. This is true hybrid, a WAN which is capable of efficiently combining high latency VSAT with low latency cellular, and any range of other bearers in-between to create an aggregate service with optimized configurable bandwidth. Failover achieves none of this, and tomorrow’s data-hungry world is rapidly going to wake up to this.

Related posts

Sidus Space Signs MOU with Exo-Space to Support Edge Computing Technology

Akebono selects Solid Edge, Teamcenter X and PROLIM for Digital Transformation

CIO Influence News Desk

CIO Influence Predictions Series: Importance of AI and Anticipated Increased Use Of Visual Technology

Dhruba Borthakur