K80 Cooling – A Liquid Cooling Use Case

Nov. 21, 2016
Liquid cooling isn’t just for servers and data center racks any longer. Now, learn about specific use-cases revolving around cooling GPU cards like the NVIDIA Tesla K80!

Did you know that graphics processing units (GPUs) are being used more for big data workloads? Did you also know that these processors have become much more powerful in the workloads they can handle and the data they can process? Even Amazon has jumped on board with making GPU compute mainstream. AWS now deploys NVIDIA Tesla K80 cards to run a variety of applications, including traditional simulation and modeling workloads typical in the HPC market as well as the training of neural networks in deep learning and GPU-accelerated databases. In this special report from Ebullient, we learn about some next-generation use-cases around Tesla K80 cooling. Most of all, you’ll learn about new types of use-cases like powerful workload processing, big data, and even business analytics applications which require the processing power of a K80 GPU card.

Before we go on… What is a Tesla K80?

NVIDIA Tesla K80 graphics accelerator are high-performance devices featuring dual GPUs for advanced and accelerated processing.

The GPU Cooling Challenge

Download the Data Center Frontier Special Report on Liquid Cooling

Organizations deploying these GPU cards have found that Tesla K80 cooling is difficult to accomplish with traditional air cooling systems. In fact, many recent inquiries and conversation points specifically revolve around administrators struggling to keep their K80 GPUs from overheating. Here’s the reality: cooling systems like those from Ebullient are not just for high-density servers and CPUs. Ebullient’s two-phase liquid cooling systems are extremely versatile and can cool everything from Intel Xeon CPUs up to high-performance GPUs, such as NVIDIA Tesla K80 graphics accelerators.  In fact, there are specific liquid cooling technologies which are designed to work with the NVIDIA Tesla card. Ultimately, this helps optimize graphics processing, creates K80 cooling best practices, and supports advanced big data workloads. This means that administrators don’t have to create complex custom cooling solutions; and should be looking at liquid cooling for their big data and next-gen GPU requirements.

Where New Liquid Cooling Systems are Revolutionizing Data Center Design

The idea isn’t just to deploy data center liquid cooling solutions. Ultimately, data center administrators want to do this efficiently; and as easily as possible. When we look at Ebullient and their cooling system, we see that it installs without any permanent modification to most server designs. This includes standard servers, and blades; as well as Open Compute Project (OCP) servers like those used by Facebook. Ebullient cooling modules replace the standard processor heat sinks and are clamped  in place through existing motherboard mounting holes. From there, an Ebullient-equipped server requires only minimal air exchange, since Ebullient cooling modules absorb a majority of total server heat by vaporizing the engineered fluid, Novec 7000.

Today and moving forward, many analysts and data center professionals agree—the next-generation of data centers will have an evolved solution around keeping servers and infrastructure cool. That solution will revolve around liquid cooling technologies. As adoption continues to grow, those data centers which take advantage of these systems earlier will create competitive advantages around the workloads they host. And, they’ll create operational efficiency all while supporting advanced business use-cases.

Download this special report today to learn how next generation data centers will overcome the inherent disadvantages around air cooling through liquid cooling technologies. Most of all, you’ll understand how, representing a paradigm shift in cooling technology, these new types of  liquid cooling platforms are revolutionizing  hardware design, development, and digital data center deployment.

About the Author

Bill Kleyman

Bill Kleyman is a veteran, enthusiastic technologist with experience in data center design, management and deployment. Bill is currently a freelance analyst, speaker, and author for some of our industry's leading publications.

Sponsored Recommendations

The AI Disruption: Challenges and Guidance for Data Center Design

From large training clusters to small edge inference servers, AI is becoming a larger percentage of data center workloads. Learn more.

A better approach to boost data center capacity – Supply capacity agreements

Explore a transformative approach to data center capacity planning with insights on supply capacity agreements, addressing the impact of COVID-19, the AI race, and the evolving...

How Modernizing Aging Data Center Infrastructure Improves Sustainability

Explore the path to improved sustainability in data centers by modernizing aging infrastructure, uncovering challenges, three effective approaches, and specific examples outlined...

How Modern DCIM Helps Multi-Tenant Colocation Data Centers Be More Competitive

Discover the transformative impact of modern DCIM software on multi-tenant colocation data centers, enhancing competitiveness through improved resiliency, security, environmental...


Rethinking Resilience: Preparing Data Centers for the Next AI Wave

Tim Hysell, Co-founder and CEO of ZincFive, explains why the need for a resilient power infrastructure has never been more pressing for data center operators.

White Papers


Harnessing the Power of a Multicloud Strategy

Aug. 9, 2022
In this white paper, Flexential outlines how enterprises can get to get the greatest value out of a multicloud strategy.