Meeting the economic and capacity challenges of the evolving datacenter landscape across the Caribbean & Latin America

By Andrés Madero, CTO CALA, Infinera

Datacenters in the Caribbean and Latin America (CALA) are increasing their modularity and distributed nature.

Datacenters do not exist in isolation. Like the rest of the work, datacenters require local, regional, and sometimes submarine connectivity – high-speed optical connectivity – to bind them to other datacenters and the people and applications that utilize their computing and storage resources.

As an industry, how can we evolve and adapt datacenter interconnect (DCI) solutions to support the growing capacity demands of expanding hyperscale datacenters while also supporting the diversity of smaller, modular and distributed ones?

The answer is threefold: more compact modular optical platform choices, innovations in embedded and pluggable coherent optical engines, and increased transmission spectrum per fiber pair. In short, we need to right-size initial DCI capacity and investment to match the datacenter’s day-one transmission demands while also enabling cost- and energy-efficient expansion over time.

A Chassis to Fit Every Environment

While the DCI category started with small, server-like transponders with coherent optical engines, the industry has quickly evolved to more flexible compact modular platforms with sleds that can be mixed and matched to support practically any desired functionality, as seen in Figure 1. Chassis are available in a wide range of depths to support deployment in diverse environments. These platforms can mix optical line system and transponder functionality and can be extended with multi-chassis connectivity that enables single network element manageability with easy capacity expansion. This pay-as-you-grow model enables operators to expand only when needed, matching cost and power consumption to capacity.

Figure 1: Compact modular DCI platform with mix-and-match sleds

(Caption: - SLED-BASED ARCHITECTURE - SYSTEM ON A BLADE - Pluggables).

Moving at the Speed of Light

With increased vertical integration, leading coherent optical engines are evolving in two directions simultaneously: 1) smaller, lower-power pluggables that can reach 1,000km or more and 2) embedded, sled-based optical engines with sophisticated transmission and reception technology that maximize capacity-reach and spectral efficiency.

With increasing capabilities in small 400G QSFP-DD pluggables, IP over DWDM (IPoDWDM) is starting to be realized with deployments directly into routers and switches. Basic 400ZR pluggables support DCI applications up to 120km with fixed settings and only support ethernet traffic. More advanced 400G ZR+ pluggables offer increased programmability, support for OTN and ethernet traffic, and better optical performance to support metro/regional and some long-distance connectivity. 400G XR pluggables support both high-performance point-to-point applications like ZR+ and point-to-multipoint deployments, where a single 400G pluggable optic can communicate simultaneously with multiple 100G pluggable optics in 25Gb/s increments.

Today’s embedded engines like Infinera’s ICE6 deliver 800Gb/s per wavelength at distances approaching 1,000km, and those same 800G engines can deliver 600Gb/s up to 3,000km and 400Gb/s almost everywhere in the world, including submarine spans that can measure 10,000km. But 1.2Tb/s-per-wavelength engines and beyond are emerging from development labs and will be ideal for long-distance DCI connectivity solutions.

Embedded engines are also ideal where fiber is scarce and high spectral efficiency is required. As an example, in situations where a datacenter operator is leasing fiber, embedded optical engines can reduce operational costs by maximizing the amount of data transmission over a single fiber pair to avoid leasing a second fiber pair or trenching new fiber. The power consumption of today’s 800G coherent optical engines is also drastically improved, utilizing 89% less power per bit than similar engines 10 years prior.

Putting More Lanes and Automobiles on the Highway

In most of the world, DWDM networks have only used the C-band of fiber spectrum, but webscale operators with hyperscale datacenters were early adopters of C+L-band transmission on the same fiber. Just like adding lanes to a highway, expanding the usable spectrum on a fiber delivers more capacity. But in recent years, operators have focused on getting capacity gains by using more spectrally efficient coherent optical engines with advanced modulation schemes. As these efficiency gains diminish with each successive generation, advancements in optical line system components like amplifiers and wavelength-selective switches (WSS) can cost-effectively increase the transmission spectrum from 4.8THz to 6.1THz in both the Super C- and Super L-bands, trading a small incremental line system infrastructure cost to realize a 27% gain in incremental spectrum and transmission capacity per fiber pair. Although still early, look for Super C and Super L deployments in future DCI networks.

Figure 2: Super C and Super L spectrum expansion

Datacenter construction shows no signs of abating, and while still critical and growing, massive hyperscale datacenters are giving way to an increasing number of smaller, modular, and diverse datacenters. This change will only accelerate compact modular deployments with a variety of sleds and chassis to fit different types of environments, a collection of embedded and pluggable optical engines, and increasing transmission spectrum per fiber. When combined, this solution architecture enables us to have a low entry price, reduced power consumption, and a small footprint, while also enabling flexible scalability to meet future capacity demands.

DISCLAIMER: This content is the sole responsibility of the author and does not necessarily reflect the opinion of BNamericas. We invite those interested in participating as a guest columnist to submit an article for possible inclusion. To do this, contact the editor at telecom@bnamericas.com.