Can multi-tenant data centres keep up with leaps in link speeds and the edge data deluge?

Flexible, scalable, easy-to-manage infrastructure eases high-speed migration


Sponsored Data centres (DCs) play a critical role in how communities and economies connect and thrive in 2021 and beyond. Today, DC managers have the ability to evolve the physical layer infrastructure to keep pace with the low-latency, high-bandwidth, and reliable connectivity that subscribers demand.

The surging volume of digital traffic coursing through the DC is being driven by advancements like 5G, AI and machine-to-machine communications, which all require latency in the single millisecond range and higher bandwidths. Dove-tailing with these increasing demands for more capacity, servers, switches and connectivity in the DC are each pushing the other to be faster and more cost effective.

With 100G hitting the market in 2020 and future generations of switches and servers on schedule to require 400G and 800G connections, the physical layer must also contribute to that need for higher performance to continuously optimize network capacity. One critical step for DC operators to support ongoing transitions from 400G to 800G, 1.2 Tb and beyond is to look at their physical layer cabling options.

Yet, a fast-changing technological landscape is only half the challenge. Timing is equally critical with refresh cycles now running every two to three years. With lots of moving parts, DC operators must make difficult but necessary strategic infrastructure transitions now or find it costly if they fail to get it right.

More so for multi-tenant DC (MTDC) operators when the Asia Pacific colocation market is expected to grow at a CAGR of close to 8% during the period 2019-2025. The regional MTDC demand is spurring the growth of services such as cloud connectivity, interconnection, and hybrid infrastructure services.

Role at the edge

When faced with the escalating cost of developing DCs and procuring land, customers inevitably turn to MTDCs when they choose to deploy edge DCs to meet the demands driven by the growing use of connected devices among businesses and consumers.

“In 2021, we will see continued momentum in edge-first deployments and a wave of technology innovations across the infrastructure stack to address the increased complexity of reliably scaling and orchestrating distributed infrastructure at the edge,” said Justin Dustzadeh, CTO of global interconnection and MTDC company Equinix.

Emerging 5G capabilities also require physical infrastructure that optimally extends into the edge. According to Gartner, 75 percent of enterprise-generated data will be created and processed at the edge by 2025.

As networks ramp up support for 5G and Internet of Things (IoT), “MTDCs are looking at the ways that their own infrastructure and solution offerings can better support edge applications, and the increasing need to locate more capacity and processing power closer to end users for the best return in time performance” said Alastair Waite of CommScope’s Global Data Center Market Development unit. “As they do, they need to re-evaluate the role of their data centres and the importance that they now have in this constantly evolving environment.”

The network edge, located between the end devices (or customer premise) and the core of the network, helps to support the low latency needed for 5G-enabled applications like cloud-assisted driving. MTDCs have to be flexible to accommodate a variety of customer configurations to fully capitalise on their location at the edge of the network. A successful MTDC will also be able to provide expert guidance to support their customers’ applications.

Connectivity within the MTDC’s own network must be flexible on day one and two as well. MTDCs are supporting a more expansive and diverse network ecosystem. This requires adapting to a more distributed, collaborative environment while delivering the power needed by customers to, for instance, create the augmented intelligence and machine learning algorithms that enable the data to be processed.

Profit with minimal pain

The persistent challenge for MTDC operators is meeting these rising demands while remaining profitable and bringing new tenants online quicker to accelerate revenue.

A flexible, robust and efficient infrastructure can be upgraded or adapted to changes speedily. MTDC managers will typically have to prepare for their customers’ next jump in DC speeds in their cages after a brief plateau in current speeds.

With multiple options for higher speed optics that add complexity to design, the ability to scale up a network rapidly to meet changing requirements is required. Solutions from vendors that can be shipped fast and installed quickly essentially improve critical time to service and revenues. For example, modular physical layer solutions let MTDCs scale up gradually or quickly as their customers’ circumstances dictate.

To ramp up link speeds, reduce latency and remove bottlenecks, MTDCs must develop and deploy physical infrastructure that is flexible and future ready as well as easy to manage and migrate to those higher speeds.

Both within the customer cage and the MTDC’s own backbone, key technological considerations include new transmission protocols driven by the latest generation of optical transceivers, and the capital and operating cost of supporting that trend as higher bandwidth and capacity demands drive ever-higher fiber counts. Moving up to 400G and beyond, new optical connectors like the MPO-16 will become more popular and will need to be considered in the future.

CommScope’s High Speed Migration platform built on modular, ultra low-loss (ULL) connectivity helps MTDC operators and their customers to achieve this. The SYSTIMAX ULL solutions feature embedded intelligence that leverages CommScope’s automated infrastructure management solution and iPatch intelligent cabling to monitor and manage the physical layer in real time, down to the port level. Having this level of visibility is essential to a DC manager choosing to deploy an Off-Premise Disaster Recovery in an MTDC that is in a different part of the country or region.

From higher density optical distribution frames (ODFs) and patch panels to higher bandwidth fiber, easier cable routing and pre-provisioning services, MTDC managers have the tools to reduce infrastructure deployment times without sacrificing network design, performance or reliability. This not only eases the challenge of compressed timeframes but also a shortage of skilled personnel.

Simple building blocks

Pre-terminated fiber and copper cabling design allows quick installation and connection to switches and routers. High-density, ultra high-density, and enhanced high-density fiber patch and splice panels enable MTDCs to support increasing fiber counts while keeping the fibers secure and accessible.

High-density ODFs improve cable management by providing intuitive and easy access to all cables and connectors for rapid routing and connection, turn-ups, and maintenance. Further optimising space while maintaining ease of use and availability is the flexible fiber raceway system Fiberguide®, that both protects and helps to manage complex and dense fiber network. Its tool-less design reduces installation time and expense whilst maintaining the proper fiber bend radius.

Aligned with the need to simplify design, interchangeable 8-, 12- and 24-fiber MPO modules optimise fiber configurations while performance assurance network tools help to establish links that support any application.

The bottom line is that planning and installation of an efficient and simplified cabling infrastructure can help reduce time to market and boost ROI. The cabling and connectivity solutions must enable a faster, more reliable network that provides consistent performance throughput from the core through to the furthest edges of the network.

Sponsored by CommScope®

SUBSCRIBE TO OUR WEEKLY TECH NEWSLETTER


Biting the hand that feeds IT © 1998–2021