5 Ways Cabling & Connectivity Impact Data Center Sustainability
There are many other factors to consider when thinking about what makes a data center truly sustainable, regardless of when it was designed or built. Some of the decisions you make about cabling and connectivity can impact the sustainability of your data center—and here are five good examples.
1. The Type of Cabling Architecture
The type of cabling architecture you deploy—a traditional data center cabling architecture vs. a low-latency architecture—can impact power consumption and use of resources. The longer data takes to move from one location to another, the more energy and cooling required to support it.
Conventional three-tier networks typically use Spanning Tree Protocol (STP). In this setup, traffic can travel on two routes—but only one route is active at a time. If the primary path isn’t available, then traffic will be routed to the secondary path until the primary path is available once again. This approach can create backups on links that handle the heavy loads found in data center environments.
For this reason, today’s data center architecture is shifting toward the two-layer leaf-spine topology, which utilizes protocol such as Equal Cost Multipathing (ECMP). Because leaf-spine architecture is designed to reduce the number of “hops” in a network’s design, it relies on fewer aggregation switches and redundant paths between the access and interconnection switches. As a result, latency and power requirements are reduced.
2. Utilization of Rack Space
To maximize efficiency, it’s important to make the most of the space you have within each rack. If your data center’s racks are highly underutilized or poorly managed, then you’re not getting the most out of the resources you have available.
The more connectivity you can fit into a space, the less square footage needed to house it all—which translates to less energy usage and more space that can be used for other purposes.
For example, we’ve seen data centers that run their cabinets at only 40% capacity. Not only is this inefficient, but it also means they’re spending nearly twice the amount of money they need to on cabinets! While it’s also not ideal to fill a rack to 100% capacity, aiming for a number close to 90% will give you the best of both worlds: efficient use of space while still leaving room to scale.
This rack density must be balanced with cooling, of course (the more active equipment in a rack, the warmer it will be). There are many considerations to think about when it comes to data center cooling, so we’ll cover that in an upcoming blog. (Make sure you subscribe so you’re notified when it goes live!)
3. The Location of Patch Panels
To echo what we mentioned above regarding rack density, you can free up additional rack space to dedicate to compute and storage (the things IT is most concerned about) by mounting patch panels above a rack instead of installing them inside the rack.
You could also consider zero U vertical patching to accommodate fiber or copper connectivity in data center cabinets without taking up valuable rack space. These patch panels mount vertically inside the cabinet along equipment mounting rails. Not only does this help you maximize rack space to keep energy and space usage in check, but it also reduces the number of racks you must purchase in the first place.
4. Where You Mount Cross-Connects
While cross-connects can go into a rack, a wall-mount cross-connect lets you take advantage of free data center real estate: your walls. This not only lets you install passive connectivity on the wall, but it may also reduce the number of racks you need.
Aside from saving valuable rack and floor space, wall-mount cross-connects offer another benefit, too: When networking equipment is upgraded every three to five years—and changes in terms of form factor and density—having all equipment ports replicated on a wall-mount system ensures a smooth transition and avoids forklift upgrades to cabling infrastructure.
5. The Layer 0 Solutions You Choose
Investing in cabling and connectivity that will last for multiple generations of compute and storage devices is another way to improve data center efficiency and sustainability.
By selecting high-performance components designed to support generation after generation of IT gear and technology, you’ll save money by not having to rip out and replace your cabling infrastructure every few years. You’ll also generate less waste.
With a fiber infrastructure, for example, to boost performance and efficiency, you don’t need to swap out your cable: Just change the transceivers. When it’s designed properly, you won’t need to touch your layer 0 infrastructure when it’s time to support a new technology. Just swap out your IT gear, and you’re good to go.
Need help making changes to your cabling and connectivity to make better use of power and space? We can help you get started. Send me a note with your questions!