Deploying AI Models at the Edge

Liquid cooling at “The Chasm,” part two: The pros and cons of early adoption

Liquid cooling at “The Chasm,” part two: The pros and cons of early adoption

By Linn Gowen, VP of Business Development 1623 Farnam

In part one of this two-part series, we discussed the migration path of liquid cooling’s early adopters for data center environments in the context of Technology Adoption Life Cycle theory. In part two, we will take a closer look at the benefits and challenges edge data center operators face when deploying liquid cooling to answer the question: what will finally push liquid cooling past the chasm from early adopters into the majority?

Liquid Cooling: Barriers to Entry

The case for liquid cooling deployments in edge data centers is starting to gain more momentum. The simple reason operators are making the switch from traditional air cooling is the significant increase in efficiency. So, what is stalling the proliferation of liquid cooling throughout the industry? Let’s look at some of the reasons edge data centers are holding back on liquid cooling.

One significant barrier is the organizational change that needs to be handled by individual data center leadership. Since air cooling systems have been in use for decades, staff are very experienced with their service, parts, and operations. Implementing a new system means learning to deploy and maintain an unfamiliar type of equipment. Many leaders are apprehensive about tackling the internal challenges surrounding technical knowledge, training, internal processes, and their associated risks and costs. Their reluctance is compounded by the intimidation created by a more complex system, as liquid cooling systems have more components and more potential points of failure.

Another barrier is financing, as installing a liquid cooling system requires considerable capital expenditure. Although the new cooling system will save the data center money, many leaders get hung up on questioning just how much money and over what period. It’s the function of a short-term mindset — if the payback analysis reveals that the savings will be too far into the future, then the gains are deemed too small to justify the costs.

Of course, these barriers only apply to a retrofit in a legacy data center. For greenfield builds with no obligation to overhaul an existing legacy cooling system, optimizing around liquid cooling presents far fewer of these obstacles. New construction can ensure aspects like structural floor loading and ceiling heights already facilitate liquid cooling. Since new data centers have a lifespan of 20 to 30 years, maybe it’s only a matter of grandfathering out old technologies. If every new edge data center features liquid cooling, air cooling will eventually fade away. Is it only a matter of father time slowly phasing out old technology with old facilities, or is there an imperative for existing edge data centers to make the change, at least toward a hybrid system with some air cooling and some liquid?

Liquid Cooling: Advantages to Embracing the Inevitable

There are benefits to being part of the early majority. First of all, IT systems will only become hotter and denser as time goes on, moving closer to the day when liquid cooling is compulsory for functional reasons. Second, real estate prices and construction costs will only become more expensive, making the size of a physical footprint more important. Third, the day may come when regulatory requirements mandate high-efficiency cooling systems and reduced water usage, favoring those prepared for such policies ahead of time.

Perhaps the most significant reason to become part of the early majority is the simplest of market forces: customer demand. If we consider that liquid cooling is at the notorious ‘chasm’ stage of technology adoption, then the liquid cooling question for edge facilities is one of when and not if. So, the clock is ticking down to the day that liquid cooling becomes a given — a matter of course for data center customers. Once this happens, data centers at the edge will automatically discount themselves at the client proposal stage if they lack a liquid cooling feature, losing business by not offering accepted best practices.

In this light, if edge data centers aren’t planning for their operations to include liquid cooling at some point, they may already be behind the curve. As liquid cooling sits at the chasm, each edge data center must choose how to proceed: lead, or follow?

Read Part 1 here: Liquid cooling at “The Chasm”: Edge data centers can benefit from hyperscale technology.

About the Author

Linn Gowen is VP of Business Development for 1623 Farnam. Prior to joining 1623 Farnam, Gowen held senior sales positions at Verizon, CenturyLink/Level3, Windstream, AT&T among other companies.

DISCLAIMER: Guest posts are submitted content. The views expressed in this post are that of the author, and don’t necessarily reflect the views of Edge Industry Review (EdgeIR.com). 

Article Topics

 | 

Comments

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Featured Edge Computing Company

Edge Ecosystem Videos

Automating the Edge

“Automating

Deploying AI Models at the Edge

“Deploying

Latest News