Subscribe to Our Digest
New pilot program enables organizations to reduce reliance on centralized cooling support infrastructure by localizing fluid distribution, pumping, and thermal management at the rack level.
Airsys has announced the next generation of LiquidRackTM, a rack-level liquid cooling platform engineered to simplify deployment for mid-density AI, data center, telecom, and edge environments. The company is currently accepting candidates for pilot deployments for this advanced liquid cooling solution.
As operators move beyond the limits of traditional air cooling, many face a gap between incremental upgrades and infrastructure-heavy liquid cooling systems. LiquidRack addresses this need with a practical architecture that delivers increased rack density while streamlining how liquid cooling is deployed, scaled, and maintained. Unlike immersion liquid cooling approaches that put multiple servers into a shared tank, or direct-to-chip liquid cooling that relies on complex piping and expensive capital investment, LiquidRack integrates fluid distribution, pumping, and control directly at the server level. This key feature allows hazardous free, maintenance friendly and seamless upgrading from air to liquid cooling at very affordable cost.
“Most data center operators are looking to a practical, flexible and seamless path from air to liquid cooling to meet the challenge of high density, stranded power and ESG goals,” said Yunshui Chen, Founder and CEO of Airsys. “LiquidRack provides that path. By integrating key cooling functions at the rack, we simplify deployment and enable customers to scale performance within the constraints of their existing infrastructure.”
Designed for high-temperature operation with dry coolers, LiquidRack supports compressor-less architectures that reduce mechanical complexity, lower energy consumption, and eliminate water usage while enabling higher rack density.
Supports 0.5-8 kW per server and up to 80 kW per rack, enabling higher compute density across data center, telecom, and edge environments.
Compatible with dry coolers, adiabatic, and traditional chiller-based systems, enabling deployment in existing facilities or integration into advanced rack-level architectures for AI factory environments.
Localizes fluid distribution, pumping, and thermal management at the rack while utilizing spray-based server-level cooling to deliver targeted heat removal from the CPUs and GPUs, improving thermal efficiency and maintaining consistent performance.
Uses up to 80% less dielectric fluid than immersion systems, minimizing fluid handling and operational complexity, while a cassette-based architecture supports retrofit and new-build environments with scalable, serviceable, and incremental deployment without large-scale facility redesign.
By reducing cooling overhead and localizing thermal management at the rack, LiquidRack enables operators to reclaim stranded capacity in retrofit environments and maximize the amount of provisioned power allocated to IT compute across greenfield and modular deployments.
LiquidRack is designed for operators transitioning beyond traditional air cooling who need higher density without introducing unnecessary complexity. Rather than requiring large, centralized liquid cooling infrastructure, the system integrates key cooling functions directly at the rack, enabling a more practical and incremental approach to liquid cooling adoption.
This allows operators to increase compute density within existing facilities, deploy liquid cooling in a modular fashion, and scale over time without committing to full facility redesigns or highly specialized architectures. The result is a solution that aligns with how most data centers evolve—step by step—while still supporting the growing thermal demands of AI and high-performance workloads.
“Not every AI deployment looks like a hyperscale training cluster,” said Alex Cordovil, Research Director at Dell’Oro Group. “A meaningful share of enterprise and edge workloads is expected to sit within the 40–80 kW-per-rack range, inside facilities that were never designed for liquid cooling. Rack-level liquid cooling architectures can offer a pragmatic path to higher densities without the infrastructure overhaul that full retrofits require.”
LiquidRack is designed to support a wide range of deployment models — from traditional data centers to edge and telecom environments. Its rack-level architecture enables self-contained, modular cooling deployments that align with both centralized facilities and distributed infrastructure. In modular edge deployments and future telecom environments, where space constraints, deployment speed, and operational simplicity are critical, LiquidRack provides a practical path to higher-density compute. By localizing fluid distribution, pumping, and control within the rack, the system reduces reliance on centralized infrastructure while maintaining scalability and serviceability across diverse deployment scenarios.
By consolidating cooling functionality within the rack, LiquidRack reduces the need for centralized liquid cooling support systems and associated infrastructure. This approach enables faster deployment timelines, reduced mechanical and operational complexity, and modular, incremental scaling aligned with capacity growth. The platform’s 2U cassette-based design, integrated pumping, and serviceable architecture further support simplified maintenance and operational continuity.
As power availability becomes a primary constraint in modern data centers, operators are increasingly focused on how infrastructure design impacts usable compute capacity. This shift is reflected in the concept of Power Compute Effectiveness (PCE).
As noted by Dr. Rand Talib, Research Analyst at Uptime Institute: “PCE is a metric developed by cooling system provider Airsys to provide greater transparency into power allocation within a data center’s provisioned electrical envelope……….Cooling architecture decisions directly influence how much of the provisioned power envelope is allocated to facility systems and how much remains available for IT compute.”
By reducing cooling overhead and distributing thermal management at the rack level, LiquidRack supports more efficient allocation of provisioned power to IT compute.
To be considered for pilot deployments of LiquidRack, contact Airsys
Airsys will be exhibiting at the Data Center World conference in Washington, DC fromApril 20-23, 2026 at booth #1016.
More information about LiquidRackTM from Airsys