How carrier hotels have evolved into edge facilities for digital interconnection

Carrier hotels have evolved over time as the number of long-distance network carriers has increased and the transition of voice traffic to data has evolved. These facilities are now considered “edge” facilities and allow network providers to connect to each other’s backbone without having to duplicate long-haul infrastructures, bringing services closer to where people need them. They include meet-me-rooms (MMRs), which allow providers to physically connect to other networks. Consumers benefit from a cost-effective, single point of entry to a suite of network providers.

In the context of edge computing, carrier hotels become essential due to their ability to facilitate low-latency connections. Edge computing relies on proximity to end-users, and carrier hotels, often situated in metropolitan areas, serve as ideal locations for deploying edge infrastructure. These facilities allow edge computing providers to establish connections with multiple carriers, ensuring diverse and reliable network paths for data transmission.

“An edge data center is a data center located close to the edge of a network. The concept is based on the distributed architecture of edge computing in which data is processed in the closest proximity possible to the originating source rather than relying solely on a centralized cloud or data center,” says Hallberg.

“Due to proximity to end users, edge data centers deliver lower latency and faster processing times, resulting in improved overall experiences for applications that require low latency or high bandwidth and where real-time processing is crucial, such as in Internet of Things (IoT) devices, autonomous vehicles and augmented reality applications. Put simply, the shorter the distance an organization’s data must travel to users, the faster it can reach its desired audience. Geography is the only thing that solves latency.”

Carrier hotels evolution

Carrier hotels are rooted in the original network—telephone exchanges, “Central offices.”

Hallberg mentions that generations ago, the only way that one could call a person in another area was to transfer the traffic from one network route to another route via an operator physically switching the traffic between nodes to interconnect.

“Divestiture of AT&T in 1984 resulted in more service providers for various telecommunications services. The resulting competition required that telephone companies establish their infrastructure in these common locations within major cities so the voice traffic could be transferred efficiently between their network routes as well as other providers’ routes. With innovations and changes in the industry landscape, hub and spoke, network to network interfacing laid the very foundation for telecommunications we still rely on today and ultimately led to what we now call carrier hotels,” she adds.

“A carrier hotel is a building strategically based in a downtown location that houses networks and cloud services. An enterprise can colocate (or connect to) to these providers and cloud services from this building which gives them easy access to all of the network connections in that carrier hotel.

“Interconnection is a word thrown around the data center industry a lot. However, what it boils down to is choice. The choice to connect to any network or cloud provider enables businesses to scale rapidly while reducing time, CAPEX and mitigating risk. Choosing a carrier hotel is the first step to building an interconnection strategy.”

Meet-me-rooms?

Within carrier hotels, meet-me rooms are dedicated spaces where different carriers physically connect their networks. These rooms house the necessary infrastructure, such as fiber optic cables and network switches, to enable seamless interconnection between diverse networks. Meet-me rooms act as neutral grounds where carriers can exchange traffic efficiently.

Hallberg says that giving an enterprise freedom to choose who they want to connect to is the core value of interconnection. Once an enterprise team identifies the carrier hotel they want to be in, then it’s time to connect their infrastructure to it. This is where the meet-me-room (MMR) comes into play.

“An MMR is a centralized point, typically a dedicated secure space, inside the data center where cloud service providers, telecom carriers, internet service providers and other companies can physically interconnect their networks to establish network connections and exchange data traffic,” she explains.

“MMRs allow providers who handle a lot of online consumer traffic to connect to one another and exchange data by directly connecting to the MMR. By connecting these networks, companies can route digital traffic to avoid fees charged by local carriers for “local loops or the last mile” to the end location (connecting from the carrier presence to their office or private data center).

“No carrier has ubiquitous network coverage, so they leverage each other’s network to meet their customers’ connectivity requirements. Think of the MMR as a shortcut for providers to exchange traffic at a lower cost with their customers. It also provides a 24/7 staffed and secure point to limit the risk of downtime or physical damage of connections between providers.”

Designing Meet-Me-Rooms: Two zones

Hallberg explains that the first zone is the Main Distribution Frame (MDF) of the facility. The MDF sits outside of the customer’s colocation space, inside the data center, serving as the demarcation (DMARC) point from the carriers to the data center operators-owned or colocation facility.

“This zone is where companies will cross-connect with each other. Companies connect in the MMR, which is managed by the data center facility team for security and continuity. The MMR is under tight change control procedures to ensure modifications do not impact carrier operations,” she adds.

“The second zone is the carrier’s space where their equipment is located. This could be a dedicated suite, cage or secure cabinet. Carrier customers of the data center operators “,DSOs,” can mount and deploy their optical equipment here, where incoming services are managed and distributed to customers or other providers.

“Most often, the second zone has very high security, monitored 24/7 by the data center team. Some data centers are network neutral allowing any carrier who wants to collocate their equipment in that facility, the data center operator provides the space and power in this zone. Ultimately, this zone provides the “choice” customers are looking for, balanced with cost-effectiveness for the interconnection.”

Carrier hotels in cloud connectivity

Carrier hotels are no longer just the home of carriers and enterprises. According to Hallberg, they are now home to the largest Public Cloud Service Providers (CSPs) such as Amazon Web Services, Microsoft Azure, Google Cloud Platform, IBM and Oracle.

“Though CSPs prefer to operate efficient, purpose-built compute facilities themselves, they recognize the need to establish their network nodes, called onramps, within locations their consumers can easily access. Onramps located within the data center provide companies within that data center the lowest latency possible,” she explains.

“A CSP selects data centers to deploy their onramps in facilities with rich ecosystems of carriers and networks that serve end-users or consumers of internet services that connect directly to residential or business customers. Colocation providers of carrier hotels serve as trusted partners to CSPs, providing onramps for their cloud or hybrid-cloud implementations via software-defined networking and interconnection platforms.”

While the flexibility of technologies, infrastructure and operation of carrier hotels have certainly evolved since the central telephone exchange days, the core concept has remained: to serve as a hub that enables providers to serve their customers.

Internet ecosystem and global connectivity

The cost benefit is the major differentiator in a carrier hotel, Hallberg notes.

“Instead of an organization having to procure connectivity to every provider they want to connect to, they can now simply collocate equipment and connect their infrastructure into any provider within the carrier hotel, allowing flexibility to scale network providers and capacity up and down, increase network performance (lower latency), increase network consistency, and directly connect to internet exchanges and cloud platforms,” she explains.

“The carrier hotel is the foundation of interconnection. It helps providers find a common meeting place to extend their reach, while helping businesses to connect to them quickly and easily.”

As the world’s data deluge continues at astronomical levels, the importance of the vast interconnected cloud and carrier partner ecosystem transporting that data cannot be overstated. From data centers to cloud service providers to wireless and wireline network providers, each player in this communication chain plays a significant role in moving data around the globe.

“But none, perhaps, holds a more pivotal role than subsea cables. With 95% of international internet traffic flowing through cables laid on the ocean floor, these underwater links are literally connecting the world and are the most critical endpoints in an organization’s edge networking strategy,” says Hallberg.

“Powered by fiber optic technology encased in tubing, subsea cables serve as underwater connection corridors that transmit data between land masses and continents at lightning-fast speeds. Underwater cables connect with land-based network infrastructure at landing stations that feed into carrier hotels.

“Carrier hotels are the Grand Central Stations of interconnectivity, housing network infrastructure for a plethora of telecom and IT service providers and are generally regarded as among the most highly interconnected of data center locations. Many of these carrier hotels are located at the edge of major population centers to allow data exchange between businesses and end users at the edge.”

Further advancements in technology

The synergy between edge computing and carrier hotels/meet-me rooms becomes evident when considering the requirements of edge applications. Edge computing demands proximity to end-users to reduce latency and enhance user experience. Carrier hotels, strategically located in urban areas, provide the necessary connectivity to serve these localized demands.

“As AI continues to evolve, both AI solution providers and enterprise users of AI will drive even more computational power, which will require more servers in the data center as well as low latency, scalability, security and reliability. AI needs will not only drive increased capacity, but also a collaborative ecosystem,” continues Hallberg.

“Edge data centers and carrier hotels will play a critical role – as they always have – in meeting these evolving demands through cloud onramps, network connectivity and providing the physical space and power required for each customer. Network-central and network-neutral edge data centers and carrier hotels are especially well equipped to meet these demands at scale.”

Hallberg predicts that technology advancements like AI and 5G, will continue to drive greater demand for computing power and capacity.

According to a recent research published by the Data Center Coalition, while data center computing workloads increased nearly 550 percent between 2010 and 2018, electricity consumption grew only six percent, due to efficiencies from cloud migration, improvements in cooling systems and other operational and technological efficiency gains at modern data centers.

“This means more power will be needed to operate data centers and carrier hotels. The good news is that the data center industry has been making great progress in sustainable growth,” she concludes.

Carrier hotels and meet-me rooms form the backbone of connectivity for edge computing. Their strategic locations, neutral interconnection environments, and scalability make them pivotal in the deployment of edge infrastructure.

Article Topics

AI  |  carrier hotel  |  Cologix  |  connectivity  |  edge computing  |  MMR