What is Direct Liquid Cooling?
Servers have become essential in our life, particularly since institutions, companies and consumers use the Cloud daily. The number of data centers keeps increasing. Servers also become more and more powerful and therefore, generate more and more heat. While still widely used in data centers, the potential of traditional air-cooling systems is limited, more so when taking the impact on the environment into account. A few technologies based on liquid emerged, among them is direct liquid cooling (DLC), also called direct to chip liquid cooling.
How does DLC work?
The direct liquid cooling process typically works in the following order:
Cold plates mounted on processors and other power-hungry equipment absorb excess heat, preventing them from overheating.
The liquid transfers through the coolant distribution unit (CDU) into a heat exchanger.
The heat exchanger cools the liquid through air exposure or other methods, allowing it to be used again. At this point, heat is expelled from the circuit.
The liquid is sent through the cooling loop again, ready to remove heat from other components.
Benefits and Applications of Direct-to-Chip Liquid Cooling
This advanced form of liquid cooling offers several key benefits, especially in data center environments where processor chips can be prone to heat exposure:
Enhanced Cooling Efficiency: By directly applying coolant to the heat source, this method achieves more efficient heat transfer compared to air cooling. This efficiency is crucial for high-density data centers where air cooling might not suffice.
Energy Savings: Liquid cooling systems typically require less energy to operate than air cooling systems. This reduction in energy consumption translates into lower operational costs and a smaller carbon footprint for data centers.
Improved Performance: Components cooled by liquid systems can maintain higher performance levels since they are less likely to overheat. This improvement is particularly beneficial for applications requiring sustained high performance, such as scientific computing and large-scale simulations. Additionally, CPUs can run at their “boost” speeds longer than with air-cooled systems because liquid cooling keeps the CPU temperatures lower for extended periods, preventing thermal throttling and maintaining peak performance.
Space Optimization: Direct-to-chip liquid cooling allows for higher computational density, meaning fewer racks or space are needed to achieve the same compute power as compared to air-cooled systems. This space efficiency is crucial in environments where space is at a premium, allowing data centers to maximize their compute power within a smaller footprint.
Reduced Noise: Liquid cooling systems operate more quietly than air cooling systems because they rely less on large, noisy fans. This noise reduction can be a significant advantage in environments where noise levels are a concern.
Cooling Challenges for AI, HPC, and Cloud Computing
Following the booming numbers in cloud computing services, artificial Intelligence (AI) has also started to become mainstream in every part of our daily lives. Combining the high-performance computing (HPC) needed in the field of scientific research, data centers worldwide continue to grow in number and scale. As the demand for computing power rapidly expands with limited space to be utilized, data centers are forced to come up with a much higher density solution, bringing unprecedented loads to the traditional air-cooling system design. Furthermore, the vibration generated by cooling fans in a server may result in damage to server components such as HDDs, and the noise could potentially damage the hearing of workers that routinely maintain IT hardware.
It is estimated that data centers now consume approximately 1% of the world’s electricity, with the average power usage effectiveness (PUE) being in the range of 1.4 to 1.6. This number indicates that around 33% of the facility power is spent not on powering the IT hardware itself but on facility costs, including a majority part on air conditioning. DLC is a much more efficient way of transferring heat away from electrical components, and since air cooling equipment within a server, such as large fans and heat sinks, is no longer required, liquid cooling systems can fit into a more compact design, potentially supporting a much greater density of CPUs and GPUs in limited space.
Overall, the popular and standardized DLC technology has been proven effective throughout the past few years, and it is now a mature technology suitable for all scales of applications from small-sized data centers and smaller HPC cluster deployments to exascale computing facilities.
What is Direct Liquid Cooling?
Servers have become essential in our life, particularly since institutions, companies and consumers use the Cloud daily. The number of data centers keeps increasing. Servers also become more and more powerful and therefore, generate more and more heat. While still widely used in data centers, the potential of traditional air-cooling systems is limited, more so when taking the impact on the environment into account. A few technologies based on liquid emerged, among them is direct liquid cooling (DLC), also called direct to chip liquid cooling.
