Posts

By Roopashree Honnachari, Industry Director for Business Communication Services & Cloud Computing at Frost & Sullivan

In our ongoing research here are Frost & Sullivan, we tend to think about Ethernet and Internet Protocol (IP) in terms of watershed moments: The two standards came together to replace the disparate networks businesses once relied on to enable communications and data transfer. Before Ethernet and IP, the communications network (phones connected to a central call control unit or PBX), the security network (cameras and access control connected to a dedicated system) and the data network were delivered over their own dedicated infrastructure and cabling. Ethernet and IP offered the promise of a single network on which all systems can interoperate, while also delivering the economies of scale that come with a shared infrastructure.

Unfortunately, several unintended consequences have taken some of the shine off the idea of an integrated network. Rather than serving as a beacon of interoperability, today’s business networks can quickly add to the workload and stress of network administrators. Each of the once-discrete networks comes with its own unique set of requirements, which are often at odds with one another. In dealing with conflicts, many admins find themselves tweaking and readjusting network settings or creating a complex set of virtual LAN (vLAN) policies. As the name suggests, the vLAN approach divides the physical network into a series of logical ones; it’s an administrator-intensive effort, devices often land in the wrong place, slowing performance, as well as increasing costs and network complexity.

In addition, ongoing cyber threats to the entire network remain, meaning that rogue apps, malware or denial of service attacks have the potential to take down not only PCs and business applications but also critical security and communications devices. Likewise, with more and more IoT endpoints connected to the network, new attack vectors are available to hackers to disrupt all of a business’ connected devices.

As Frost & Sullivan analysts across several research practices came together to develop a new set of best practices for local area networks, we factored both ongoing security threats as well as network simplification into what we call the “Modern LAN principles.” To address both the opportunity and challenges of network convergence, we included this recommendation:

When possible, construct physically separate but functionally integrated IP network paths for different and dedicated applications, ensuring mission-critical platforms are not impacted by disruptions or intrusions of the primary business network. By doing so, organizations have the option to create separate networks or connect them on site or in the cloud with a single cable.

So, what does a “physically separate but functionally integrated” network look like? In the diagram below, you will notice that the IP camera and IPTV in the upper right corner are physically separated from the PC network in the center and the IoT sensors on the left. This has two advantages. First, management of the throughput, power and configuration needs of the camera can be managed separately from the PC network, largely eliminating the need for vLAN configurations. Second, in the event of a cyber attack on the IoT or PC network, the security network is protected by its own firewall. And if a rapid response is needed, it can be unplugged from the rest of the network and operated independently. The PC and IoT networks can also operate in an independent but integrated fashion as needed.

“Physically separate but functionally integrated” is just one of several design best practices incorporated in the Modern LAN. To learn more, read the whitepaper “The Modern LAN: Rethinking Network Design for the Modern Age”, available at https://go.frost.com/Lan

Learn more about Modern LAN enabling technology

Learn more at http://themodernlan.org/

Follow us on Social Media

By Melanie Turek, Vice President of Research, Connected Work, Frost & Sullivan

Ask most network architects about the local area networks they manage, and they will likely regale you with a clear list of their preferred networking vendors and products for each of their environments. One thing that rarely comes up: the exceptions that lurk within every corporate LAN.

Exceptions are the one-offs, such as that IP phone on the loading dock that is just a little too far away from the nearest wiring closet to connect via standard Ethernet, or using dongles and an existing coaxial cable to plug a security camera into the network. All but the most generic of local area networks contain these dirty little secrets; the network administrators and architects I’ve spoken to will admit to them only grudgingly, with a wink and a nod.

The challenge, however, is that with the convergence of technologies on the LAN—including not only communications and security devices, but a whole fleet of IoT sensors and endpoints—the risks posed by seemingly benign one-off exceptions are all too real. That casual wink and nod could easily turn into a look of dread as more and more networked devices require non-standard equipment to work as desired—and blow up any semblance of network optimization, performance and security in the process. What’s more, administrators could find themselves dealing with scores of unique fixes, without relying on the rigorous processes that are foundational to any successful IT operation.

Helping companies deal with such exceptions is just one of several challenges Frost & Sullivan’s analyst team aimed to mitigate as we developed a new set of best practices for local area networks, which we are calling the “Modern LAN.” The solution: When it comes to dealing with exceptions, it is time to stop treating them as exceptions.

As non-traditional devices become the norm, network architects simply must come to terms with the fact that solutions such as Long-Reach Power over Ethernet (LRPoE) switches must be included as part of the approved network architecture, even if that means deploying new network design techniques and technologies. It also means being open to the possibility of including new suppliers and certifying new products as part of corporate-approved solutions. Incorporating and supporting a wider array of compatible network infrastructure gear gives both architects and the administrators in the field a fresh set of tools to deploy networks reflective of the devices that are making their way into corporate offices, shop floors and remote sites.

“Dealing with Exceptions” is just one of several design best practices incorporated in the Modern LAN. To learn more, read the whitepaper “The Modern LAN: Rethinking Network Design for the Modern Age”, available at https://go.frost.com/Lan

Learn about Modern LAN enabling PoE innovations

Follow us on Social Media

By Dilip Srangan, Frost & Sullivan Global Research Director, Internet of Things (IoT) & Digital Transformation

Here at Frost & Sullivan, my colleagues and I have been looking at the conventional wisdom associated with building and supporting local area networks, and we came away with a serious question: Is the current way to do things actually the best way to support the Internet of Things (IoT)? The answer, it seems, is no.

The IoT has a significant impact on business networks, and all of the related requirements—including power, bandwidth and even where devices are deployed—are unique from anything else on the network. As a result, the way most network architects have grown up thinking about the LAN doesn’t work anymore. It’s clear that the industry needs a modern approach to network design. Modern LAN principles, developed by Frost & Sullivan, seek to develop a new set of best practices around local network design.

The first principle guiding Modern LAN design is one that particularly speaks to the challenge of incorporating IoT into a local area network: Adopt an outside-in approach to local area network design and planning. Before new endpoints are deployed, network managers must identify the power, bandwidth and application requirements of each unique physical device, in order to determine the best topology and infrastructure to support it.

Most network designs focus on the core network and move out to the devices that run on it. That delivers a homogeneous network, but it ignores the needs of the devices themselves—and the business benefits that said devices provide. If they take this approach as they embrace the IoT, businesses will likely pay for network bandwidth they don’t need, create a power over Ethernet (PoE) distribution nightmare and rack up a very large cabling bill.

The outside-in approach turns the design on its head, focusing on the devices at the end of the network matching them with the most appropriate network switches and power distribution infrastructure. With the outside-in principle applied, the network is more efficient and optimized for all the devices that connect to it. That, in turn, ensures better performance, thereby delivering the benefits and ROI promised by digital transformation and the IoT. Businesses can even save a few bucks by using existing and still useful wiring.

Outside-in is just one of several design best practices incorporated in the Modern LAN. To learn more, read the whitepaper “The Modern LAN: Rethinking Network Design for the Modern Age”.

Thank you, Dilip for your expert insight into Modern LAN design. Keep it locked on the NVT Phybridge news page for more industry analysis into the Modern LAN principles.

Follow us on Social Media