{"id":12657,"date":"2026-09-08T11:03:54","date_gmt":"2026-09-08T15:03:54","guid":{"rendered":"https:\/\/www.qats.com\/cms\/?p=12657"},"modified":"2026-09-08T11:08:11","modified_gmt":"2026-09-08T15:08:11","slug":"hybrid-cooling-air-and-liquid-double-team-heat-in-data-centers","status":"publish","type":"post","link":"https:\/\/www.qats.com\/cms\/2026\/09\/08\/hybrid-cooling-air-and-liquid-double-team-heat-in-data-centers\/","title":{"rendered":"Hybrid Cooling: Air and Liquid Double-Team Heat in Data Centers"},"content":{"rendered":"<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-full is-resized\"><img loading=\"lazy\" decoding=\"async\" width=\"600\" height=\"400\" src=\"https:\/\/www.qats.com\/cms\/wp-content\/uploads\/Title-image-4.png\" alt=\"\" class=\"wp-image-12651\" style=\"width:456px;height:auto\" srcset=\"https:\/\/www.qats.com\/cms\/wp-content\/uploads\/Title-image-4.png 600w, https:\/\/www.qats.com\/cms\/wp-content\/uploads\/Title-image-4-300x200.png 300w\" sizes=\"auto, (max-width: 600px) 100vw, 600px\" \/><\/figure>\n<\/div>\n\n\n<p class=\"wp-block-paragraph\">Much of the focus on data centers is on their very high power and cooling demands and the impact those demands have on local electrical grids and water supplies. Water is needed for thermally managing the hot-running processors, especially GPUs, at the heart of AI data center computing. Many other essential components also require cooling, but at lower levels, allowing them to be cooled using forced air.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">This is why hybrid cooling, using both liquid and air methods, is an effective way to keep hot data center components within their operating temperature ranges. Hybrid cooling enables a more balanced, incremental approach to managing different types of heat loads efficiently. Liquid cooling handles concentrated heat loads, while airflow manages the remaining thermal load.<\/p>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-full is-resized\"><img loading=\"lazy\" decoding=\"async\" width=\"748\" height=\"461\" src=\"https:\/\/www.qats.com\/cms\/wp-content\/uploads\/Fig-1-10.png\" alt=\"\" class=\"wp-image-12652\" style=\"width:626px;height:auto\" srcset=\"https:\/\/www.qats.com\/cms\/wp-content\/uploads\/Fig-1-10.png 748w, https:\/\/www.qats.com\/cms\/wp-content\/uploads\/Fig-1-10-300x185.png 300w\" sizes=\"auto, (max-width: 748px) 100vw, 748px\" \/><\/figure>\n<\/div>\n\n\n<p class=\"has-text-align-center wp-block-paragraph\"><strong><em>Figure 1 \u2013 Hybrid Air and Liquid Cooling Systems on a Data Center Server [2] <\/em><\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Forced air cools lower power components using heat sinks and air from fans and blowers. But traditional air cooling becomes inefficient as board and component densities climb, and power levels increase, justifying the need for high capacity air cooling that is not necessarily a larger fan of heat sink. &nbsp;<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Liquid cooling is provided by continuously circulating liquid coolant across the hot modules, drawing their heat away. While air cooling is provided by rows of fans, liquid cooling is a complex system of tubing and connectors, coolant distribution units (CDUs), cold plates, and coolants. [3]<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Data Center Applications<\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Let\u2019s look at hybrid cooling inside a typical server in an AI data center, where it can be one of thousands of servers running in tightly and precisely arranged rows of cabinets. Both air and liquid cooling play essential roles in thermally managing the many powered components.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Air Cooling<\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Server cases are explicitly designed to function as wind tunnels. Server airflow is generated by internal fans that create air pressure. Typically, these are high-static-pressure, counter-rotating fans that produce cooling airflow passing over and around motherboards and hot components. Fan pressure pushes heated air out of the server while drawing cool air in from outside. [4]<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">At the component level, air-cooling designs rely on conduction to move heat from a chip to a metal heat sink, followed by forced convection driven by high-static-pressure fans to carry the heat away.<\/p>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-large is-resized\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"558\" src=\"https:\/\/www.qats.com\/cms\/wp-content\/uploads\/Fig-2-11-1024x558.png\" alt=\"\" class=\"wp-image-12653\" style=\"width:631px;height:auto\" srcset=\"https:\/\/www.qats.com\/cms\/wp-content\/uploads\/Fig-2-11-1024x558.png 1024w, https:\/\/www.qats.com\/cms\/wp-content\/uploads\/Fig-2-11-300x163.png 300w, https:\/\/www.qats.com\/cms\/wp-content\/uploads\/Fig-2-11-800x436.png 800w, https:\/\/www.qats.com\/cms\/wp-content\/uploads\/Fig-2-11.png 1043w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n<\/div>\n\n\n<p class=\"has-text-align-center wp-block-paragraph\"><strong><em>Figure 2 \u2013 Fans Pull Air In and Across the Board. Warmed Air Exhausts Out the Back, e.g., into a Hot Aisle. [4]<\/em><\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Some commonly air-cooled components include:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Memory (DIMMs \/ DDR5): Cooled via linear front-to-back airflow channels passing across the motherboard.<\/li>\n\n\n\n<li>Networking (NICs \/ SmartNICs \/ OCP 3.0 cards): Low-to-moderate-power PCIe or OCP networking modules.<\/li>\n\n\n\n<li>Storage (NVMe \/ SSDs): Front-accessible U.2\/U.3 or E1.S drives.<\/li>\n\n\n\n<li>Power Supply Units (PSUs): Integrated hot-swap power supply modules equipped with their own independent, self-contained fans.<\/li>\n\n\n\n<li>CPUs: Some lower-power CPUs can be air-cooled using low-profile fan-sinks or vapor-chamber heat sinks. [5]<\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">In standard 1U and 2U multi-node enterprise server racks, CPUs and other components are almost always cooled passively using aluminum or copper block heat sinks that rely entirely on the high-RPM chassis fans located at the front of the server case.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Despite its broad applicability, air cooling is typically used for less than 20 kilowatts (kW) per rack in conventional designs, although it can reach 35 kW in exceptional cases. Anything beyond that typically requires liquid cooling. [6] Of course, the selection of a cooling system is governed by device junction temperature rather than total wattage.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Liquid Cooling<\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Liquid cooling systems circulate coolant through a network of tubing, absorbing heat from hot components, most notably high-power GPUs, many of which operate at 1,000 W.<\/p>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-full is-resized\"><img loading=\"lazy\" decoding=\"async\" width=\"698\" height=\"418\" src=\"https:\/\/www.qats.com\/cms\/wp-content\/uploads\/Fig-3-11.png\" alt=\"\" class=\"wp-image-12654\" style=\"width:554px;height:auto\" srcset=\"https:\/\/www.qats.com\/cms\/wp-content\/uploads\/Fig-3-11.png 698w, https:\/\/www.qats.com\/cms\/wp-content\/uploads\/Fig-3-11-300x180.png 300w\" sizes=\"auto, (max-width: 698px) 100vw, 698px\" \/><\/figure>\n<\/div>\n\n\n<p class=\"has-text-align-center wp-block-paragraph\"><strong><em>Figure 3 \u2013 Liquid Cooling on a Data Center GPU. [7]<\/em><\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">In data center servers, liquid cooling primarily targets ultra-high-heat-producing processors, specifically GPUs and CPUs, using direct-to-chip cold plates. In some advanced high-density 1U designs, liquid loops also extend coverage to adjacent high-power components.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Some commonly liquid-cooled components include:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>GPUs \/ Accelerators: The core AI training and inference engines that generate the vast majority of server heat.<\/li>\n\n\n\n<li>CPUs: Host processors that manage system instructions and data pipeline feeds to accelerators.<\/li>\n\n\n\n<li>Network\/AI Switches &amp; Co-processors: Specialized fabric chips, such as NVSwitches, mounted on the board to link multiple processors at high speeds. [8]<\/li>\n<\/ul>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-full is-resized\"><img loading=\"lazy\" decoding=\"async\" width=\"671\" height=\"450\" src=\"https:\/\/www.qats.com\/cms\/wp-content\/uploads\/Fig-4-13.png\" alt=\"\" class=\"wp-image-12659\" style=\"width:497px;height:auto\" srcset=\"https:\/\/www.qats.com\/cms\/wp-content\/uploads\/Fig-4-13.png 671w, https:\/\/www.qats.com\/cms\/wp-content\/uploads\/Fig-4-13-300x201.png 300w\" sizes=\"auto, (max-width: 671px) 100vw, 671px\" \/><\/figure>\n<\/div>\n\n\n<p class=\"has-text-align-center wp-block-paragraph\"><strong><em>Figure 4 \u2013 A Hybrid-Cooled 1U Rack Mount Server Chassis by Titan Rig includes 490W Cooling Capacity. A Powerfan Hub Bundle Provides Air-Cooling for Lower Power Components. [9]<\/em><\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Power of Two<\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Hybrid cars combine gas engines and electric motors to improve fuel efficiency. Similarly, hybrid cooling in data center servers combines air and liquid systems for greater thermal-management efficiency. Air cooling manages ambient and lower-density heat loads, typically up to 35 kW per server board. Liquid cooling targets high-intensity heat directly at the component level, such as CPUs and GPUs. Both systems operate simultaneously, reducing the thermal burden placed on either system alone.<\/p>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-full is-resized\"><img loading=\"lazy\" decoding=\"async\" width=\"624\" height=\"415\" src=\"https:\/\/www.qats.com\/cms\/wp-content\/uploads\/Fig-5-11.png\" alt=\"\" class=\"wp-image-12656\" style=\"width:528px;height:auto\" srcset=\"https:\/\/www.qats.com\/cms\/wp-content\/uploads\/Fig-5-11.png 624w, https:\/\/www.qats.com\/cms\/wp-content\/uploads\/Fig-5-11-300x200.png 300w\" sizes=\"auto, (max-width: 624px) 100vw, 624px\" \/><\/figure>\n<\/div>\n\n\n<p class=\"has-text-align-center wp-block-paragraph\"><strong><em>Figure 5 \u2013 Liquid Cooling from ZutaCore Closely Neighbors an Air Cooling Heat Sink on a Server Board.<\/em><\/strong> <strong><em>[3]<\/em><\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Data center cooling isn\u2019t only about installing fans and circulating coolant. Strategic design combined with optimization is crucial for ensuring that data center infrastructure and components operate at their best. This approach not only prevents devices from overheating but also maximizes the efficient use of data center resources, including power and cooling systems.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">While this post focuses on component cooling within servers, data centers also have other thermal-management systems. In particular, their HVAC systems deliver chilled airflow to targeted areas within the facility. By optimizing airflow through vents and barriers, data centers can minimize areas of stagnant air. Configuring server racks in alternating hot- and cold-aisle rows maximizes efficiency by ensuring that cooled air is delivered where it is needed most while preventing hot and cold air from mixing.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>References<\/strong><\/p>\n\n\n\n<ol class=\"wp-block-list\">\n<li>Top Image <a href=\"https:\/\/www.gigabyte.com\/Enterprise\/GPU-Server\/G262-ZL0-rev-G00\">https:\/\/www.gigabyte.com\/Enterprise\/GPU-Server\/G262-ZL0-rev-G00<\/a><\/li>\n\n\n\n<li>Tech Stories, <a href=\"https:\/\/www.techstories.co\/liquid-cooling-leak-destroys-millions-of-dollars-in-gpus\/\">https:\/\/www.techstories.co\/liquid-cooling-leak-destroys-millions-of-dollars-in-gpus\/<\/a><\/li>\n\n\n\n<li>Chatsworth, <a href=\"https:\/\/www.chatsworth.com\/en-us\/resources\/blogs\/2026\/what-is-hybrid-cooling-a-smarter-approach-to-data-center-thermal-design\/\">https:\/\/www.chatsworth.com\/en-us\/resources\/blogs\/2026\/what-is-hybrid-cooling-a-smarter-approach-to-data-center-thermal-design\/<\/a><\/li>\n\n\n\n<li>EziBlank, <a href=\"https:\/\/eziblank.com\/blog\/2022\/11\/29\/how-airflow-works-inside-of-a-server\/\">https:\/\/eziblank.com\/blog\/2022\/11\/29\/how-airflow-works-inside-of-a-server\/<\/a><\/li>\n\n\n\n<li>Cooltron, <a href=\"https:\/\/www.cooltron.com\/news\/thermal-engineer's-guide-to-choosing-ai-server-cooling-fans-2078026334647910402\">https:\/\/www.cooltron.com\/news\/thermal-engineer&#8217;s-guide-to-choosing-ai-server-cooling-fans-2078026334647910402<\/a><\/li>\n\n\n\n<li>Digital Reality, https:\/\/www.digitalrealty.com\/resources\/blog\/future-of-data-center-cooling<\/li>\n\n\n\n<li>Jetcool, <a href=\"https:\/\/jetcool.com\/post\/what-is-direct-liquid-cooling-for-ai-data-centers\/\">https:\/\/jetcool.com\/post\/what-is-direct-liquid-cooling-for-ai-data-centers\/<\/a><\/li>\n\n\n\n<li>Solidigm, <a href=\"https:\/\/www.solidigm.com\/products\/technology\/liquid-cooling-basics.html\">https:\/\/www.solidigm.com\/products\/technology\/liquid-cooling-basics.html<\/a><\/li>\n\n\n\n<li>Titan Rig, <a href=\"https:\/\/www.titanrig.com\/1u-liquid-cooled-rack-mount-server-chassis-with-490w-cooling-capacity-xpx-pro-cpu-water-block-es-guardian-powerfan-hub-bundle.html\">https:\/\/www.titanrig.com\/1u-liquid-cooled-rack-mount-server-chassis-with-490w-cooling-capacity-xpx-pro-cpu-water-block-es-guardian-powerfan-hub-bundle.html<\/a><\/li>\n<\/ol>\n","protected":false},"excerpt":{"rendered":"<p>Much of the focus on data centers is on their very high power and cooling demands and the impact those demands have on local electrical grids and water supplies. Water is needed for thermally managing the hot-running processors, especially GPUs, &hellip; <a href=\"https:\/\/www.qats.com\/cms\/2026\/09\/08\/hybrid-cooling-air-and-liquid-double-team-heat-in-data-centers\/\">Continue reading <span class=\"meta-nav\">&rarr;<\/span><\/a><\/p>\n","protected":false},"author":1001007,"featured_media":0,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[940,14,37,59,63],"tags":[924,204,980,107],"class_list":["post-12657","post","type-post","status-publish","format-standard","hentry","category-air-cooling","category-cooling","category-liquid-cooling","category-thermal-design","category-thermal-management","tag-air-cooling","tag-electronics-cooling","tag-hybrid-cooling","tag-liquid-cooling-2"],"_links":{"self":[{"href":"https:\/\/www.qats.com\/cms\/wp-json\/wp\/v2\/posts\/12657","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.qats.com\/cms\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.qats.com\/cms\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.qats.com\/cms\/wp-json\/wp\/v2\/users\/1001007"}],"replies":[{"embeddable":true,"href":"https:\/\/www.qats.com\/cms\/wp-json\/wp\/v2\/comments?post=12657"}],"version-history":[{"count":2,"href":"https:\/\/www.qats.com\/cms\/wp-json\/wp\/v2\/posts\/12657\/revisions"}],"predecessor-version":[{"id":12661,"href":"https:\/\/www.qats.com\/cms\/wp-json\/wp\/v2\/posts\/12657\/revisions\/12661"}],"wp:attachment":[{"href":"https:\/\/www.qats.com\/cms\/wp-json\/wp\/v2\/media?parent=12657"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.qats.com\/cms\/wp-json\/wp\/v2\/categories?post=12657"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.qats.com\/cms\/wp-json\/wp\/v2\/tags?post=12657"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}