Lian Li Tech

Liquid Cooling Container Data Center: Fast Deployment, High Efficiency

Discover the benefits of liquid cooling container data centers, including lower PUE, reduced operating costs, rapid installation, and flexible expansion for AI computing environments.

Direct-to-Chip Cooling Systems: Cold Plates, CDUs and Liquid Cooling Solutions

Understand the key components of Direct-to-Chip Cooling systems, including cold plates, CDUs, coolant management, and liquid cooling architecture.

Full article:

Why the NVIDIA GB200 NVL72 Demands Liquid Cooling: The New Reality of AI Infrastructure

In the race to build ever-more powerful AI systems, raw compute power has hit a hard physical wall: heat. NVIDIA’s GB200 NVL72 rack-scale system — packing 72 Blackwell GPUs and 36 Grace CPUs into a single liquid-cooled rack delivering up to 1.44 exaFLOPS of FP4 inference performance — exemplifies this shift. Its power draw reaches approximately 120–132 kW per rack, making traditional air cooling not just inefficient, but physically impossible.

This isn’t hype. It’s physics. Here’s why liquid cooling is mandatory for the GB200 NVL72, what it means in real-world data center deployments, and how organizations can prepare.

The Scale of the Heat Problem

Each GB200 Superchip (one Grace CPU + two Blackwell GPUs) has a thermal design power (TDP) around 1,200W. Across 36 such modules in the NVL72 rack, the total heat load surges past 120 kW — roughly 5–15 times what a typical high-density air-cooled rack was designed to handle.

Individual GPU dies push heat flux to 500–600 W/cm². For context, that rivals the heat density of nuclear reactor fuel rods, while standard air-cooled heatsinks manage only 5–15 W/cm² effectively.

Air simply cannot move heat away fast enough. Even with massive fans, airflow limits, thermal resistance, and noise constraints make it unworkable. Attempting air cooling at this density would require impractical fan power (potentially 20–30% of the rack’s total energy), generate deafening noise (>90 dB), and still result in junction temperatures far exceeding safe silicon limits, causing throttling or hardware failure.

NVIDIA has confirmed the GB200 NVL72 is engineered exclusively as a liquid-cooled platform from the ground up, with direct-to-chip (DTC) cold plates on GPUs, CPUs, and even NVLink switches.

How Liquid Cooling Works in the GB200 NVL72

Direct-to-chip liquid cooling circulates a specialized coolant (often water-glycol mixtures) through cold plates mounted directly on the hottest components. Heat transfers efficiently into the fluid, which is then pumped to a Coolant Distribution Unit (CDU) — either in-rack, in-row, or facility-scale — where it exchanges heat with a facility water loop or air via heat exchangers.

Typical parameters for the NVL72:

Coolant inlet around 25–45°C, exiting ~20°C warmer.

Flow rate: approximately 2 liters per second per rack.

Hybrid setups often handle the bulk via liquid (~70–80%) with some rear-door or perimeter air cooling for the rest.

This approach leverages liquid’s vastly superior thermal properties: water has about 3,500 times the volumetric heat capacity of air, enabling compact, high-efficiency heat removal with far lower energy overhead.

Real-World Performance and Efficiency Gains

NVIDIA reports dramatic improvements with the liquid-cooled GB200 NVL72 compared to traditional air-cooled H100-era systems:

25x better energy efficiency for the same power envelope.

300x better water-use efficiency by reducing reliance on evaporative cooling and chillers.

Up to 40x higher revenue potential per rack and 30x throughput due to massive density and NVLink interconnects (130 TB/s within the rack).

In practice, this means hyperscalers and AI factories can pack far more performance into the same footprint while slashing operational costs. A 50 MW data center could see millions in annual savings on power and cooling. Liquid cooling also supports higher ambient inlet temperatures, easing facility demands in varied climates.

For enterprises, this translates to faster training and inference of trillion-parameter models in a single coherent NVLink domain, reducing latency and communication bottlenecks that plague multi-rack air-cooled setups.

Deployment Challenges and Practical Considerations

Adopting GB200 NVL72 isn’t plug-and-play for most facilities:

Power Infrastructure: Racks draw 120+ kW (with peaks higher). Facilities need robust PDUs, busbars, and higher-voltage feeds (e.g., 415V/480V). Plan for redundancy and peak electrical design power (EDPp) up to 1.5x TDP.

Cooling Retrofits or New Builds: Many data centers require CDUs, manifolds, leak-detection systems, and facility water loops. Liquid-to-air solutions exist for facilities without chilled water, using sidecar or in-row units.

Space and Weight: A fully loaded rack can weigh over 1,300 kg (nearly 3,000 lbs). Structural reinforcements and rear extensions for cabling/manifolds may be needed.

Maintenance and Reliability: Modern systems include blind-mate connectors, redundant pumps, and negative-pressure designs to minimize leak risks. Partner with experienced vendors (e.g., Vertiv, HPE, Supermicro) for reference architectures that speed deployment by up to 50%.

Helpful Tip: Start with a site assessment focusing on power density, existing chilled water capacity, and PUE goals. Hybrid liquid-air designs can bridge gaps during transitions. Budget 4–8 weeks for cooling infrastructure in retrofits.

Why This Matters Beyond the Data Center

The shift to liquid cooling isn’t just technical — it’s economic and environmental. AI demand is exploding, but so are energy and water concerns. Liquid cooling helps data centers become more sustainable by lowering overall Power Usage Effectiveness (PUE often approaching 1.1–1.2) and reducing water consumption dramatically.

For businesses deploying AI — whether training custom models, running inference at scale, or building private clouds — choosing liquid-cooled platforms like the NVL72 future-proofs infrastructure. It enables higher density, better performance-per-watt, and lower total cost of ownership despite higher upfront costs.

The Road Ahead

NVIDIA’s trajectory with GB300 and beyond points to even denser systems. Liquid cooling is now table stakes for flagship AI hardware. Vendors are innovating with immersion options, advanced CDUs, and modular “AI factories” to make adoption easier.

Organizations ignoring this shift risk being outpaced. Those who invest wisely — partnering with OEMs offering turnkey liquid-cooled solutions — will unlock the full potential of next-generation AI while controlling costs and environmental impact.

NVIDIA GB200 NVL72 Liquid Cooling Solution

Lian Li offers an integrated solution covering system design, data center engineering and construction, equipment R&D and manufacturing, global installation, and full-lifecycle O&M. Our portfolio—ranging from cold plates, rack-mount and cabinet-style CDUs, and liquid-cooled server cabinets to containerized data centers—is deeply optimized for the NVIDIA GB200 NVL72, enabling a PUE of below 1.15.

Bottom line: The NVIDIA GB200 NVL72 must use liquid cooling because the laws of physics leave no alternative at this performance level. Far from a drawback, it’s a gateway to unprecedented efficiency, density, and capability. For forward-looking data center operators and AI leaders, embracing liquid cooling isn’t optional — it’s the smartest path to staying competitive in the AI era.

How to Choose a Liquid Cooled Server Rack for AI Infrastructure

Why AI Data Centers Are Switching to Cold Plate Liquid Cooling Solutions

Why Liquid Cooling Is the Future of Bitcoin Mining: Antminer S23 Explained

Discover how hydro cooling technology improves Bitcoin mining efficiency, rack density, and operational stability. Learn how Antminer S23 miners support next-generation mining farms.

Why Liquid Cooling Orders Are Set to Surge in Q3 as Hyperscalers Deploy $725 Billion in AI Infrastructure

$725 billion. That’s the staggering scale of AI infrastructure capital expenditure planned by the big four U.S. tech giants in 2026—Microsoft (~$190B), Amazon (~$200B), Google (~$175-185B), and Meta (~$115-135B). This figure surpasses the entire global semiconductor industry’s annual revenue (around $600 billion) and equates to burning through roughly $230,000 every second. While Q1 2026 saw these companies spend about $130 billion, primarily on GPUs and site preparation, the critical liquid cooling segment—the essential thermal management layer for next-gen data centers—is poised for dense order fulfillment in Q3, not earlier quarters.

This timing isn’t speculative; it’s dictated by hard supply chain physics. GPUs typically have a 3-6 month lead time from order to delivery, while data center construction and fit-out take 6-9 months. GPUs ordered in Q1 arrive in Q2-Q3, and facilities readied in Q2 require advanced cooling systems commissioned in Q3. These timelines converge in the third quarter, creating a natural peak for liquid cooling deployments.

From Optional to Mandatory: GPU Power Demands Shatter Air Cooling Limits

Liquid cooling has shifted from a premium option to a non-negotiable requirement because chip power consumption has pushed air cooling to its physical limits. Consider the progression:

H100: ~700W — air cooling barely suffices

B200: 1000W — pushing air cooling’s theoretical ceiling

GB300: 1400W — air cooling is no longer viable

Rubin Ultra: 2300W — more than double the power of a typical microwave

At CES 2026, NVIDIA confirmed that its Rubin platform will be 100% liquid cooled, eliminating traditional fans in favor of advanced microchannel cold plates, manifold systems, and high-capacity (1MW+) Coolant Distribution Units (CDUs) in its fourth-generation architecture. Google’s TPU v7 and v8 platforms have similarly mandated liquid cooling, with near-100% adoption expected. UBS forecasts strong TPU shipments, driving substantial liquid cooling demand from Google alone.

U.S. policy and corporate standards are accelerating the shift. Major cloud providers now require liquid cooling for high-density racks (e.g., 21kW+), reflecting both physics and efficiency mandates.

Q3 Order Concentration: Hyperscaler Delivery Timelines Align

Q3 represents a concentrated wave of liquid cooling deployments rather than scattered activity.

On the U.S. hyperscaler side, Vertiv stands out as a major beneficiary. The company is positioned for significant CDU and system wins with Google, NVIDIA GB300/Rubin-related projects, and Meta. Google’s supplier events and NVIDIA’s Rubin production ramp make Q3 a pivotal window for converting backlog into recognized revenue. Vertiv’s liquid cooling portfolio has expanded aggressively to support direct-to-chip solutions for AI workloads.

Domestically focused cloud and AI builders are also accelerating. Major players like Microsoft and Amazon are scaling AI data centers with high liquid cooling penetration. Headcount growth in liquid-cooled server procurement is robust, with plans to push adoption well above 30% in new builds. AI training clusters already show high liquid cooling rates (70%+), with overall server penetration climbing toward 50%.

Slicing the $725 Billion Pie: Where Liquid Cooling Captures Value

Liquid cooling typically accounts for 10-15% of data center infrastructure spend. Applied to the $725B total, that implies a $72.5-108.7 billion addressable opportunity (though not all flows exclusively to new liquid systems, as legacy air-cooled capacity persists). Global liquid cooling market projections show strong growth: around $13.8 billion in 2026, expanding to over $21 billion in 2027 (CAGR ~57%). North American demand, driven by hyperscalers, leads the charge.

Per-rack value is rising fast. GB300 liquid cooling suites run ~$97,500 per rack, while Rubin Ultra configurations (with upgraded microchannel plates) climb to $121,300+—a ~24% increase. Higher power density directly translates to greater system complexity and revenue per unit.

Key U.S. supply chain players and localization trends:

CDUs: Vertiv leads with strong hyperscaler qualifications and high margins on advanced units. Other players like LiquidStack (Trane) and Accelsius are gaining traction.

Cold Plates: Boyd Corporation, CoolIT Systems, and nVent hold strong positions, with ongoing yield improvements on high-performance microchannel designs pushing margins from low double-digits toward 25%.

Quick Disconnects (UQD): U.S. and allied suppliers are advancing certifications for NVIDIA platforms, targeting higher domestic content.

Google’s TPU ecosystem represents one of the largest single-customer growth drivers, with potential for outsized order elasticity in 2027.

Takeaways for Stakeholders

For liquid cooling manufacturers: Q3 is the proving ground for capacity, quality, and certifications. Focus on flawless execution for hyperscaler audits—consistency wins long-term share. For those outside major qualification pipelines, prioritize NVIDIA Rubin, Google TPU v8, and equivalent platform validations now. The 12-18 month cycle means missing this window delays revenue into 2028+.

For investors: Watch Q3 earnings for confirmation signals—strong sequential overseas/U.S. hyperscaler revenue growth (Vertiv as a bellwether), cold plate margin recovery, and new platform certification wins. Per-rack value uplift provides built-in revenue leverage even at flat volumes.

For data center operators: In the liquid cooling era, vendor selection shifts to differentiation. Prioritize:

Suppliers with full platform certifications (NVIDIA GB300/Rubin, Google TPU, etc.)

Integrated, end-to-end solutions with clear system-level accountability

Strong Service Level Agreements (SLAs) covering full lifecycle costs, where operations can represent 20-30% of total TCO

The $725 billion AI buildout is reshaping the compute stack from the ground up. Liquid cooling’s Q3 inflection isn’t hype—it’s the inevitable result of power curves, delivery timelines, and engineering reality meeting at scale. Companies positioned in the right part of the stack stand to capture durable, high-margin growth through the rest of the decade.

Quick Tip: As AI workloads continue to grow, efficient cooling has become critical for high-density data centers. Lian Li Liquid Cooling delivers complete liquid cooling solutions covering the entire ecosystem—from cold plates and CDUs to liquid cooling racks and containerized data centers. Powered by its own R&D team and Shenzhen-based manufacturing capabilities, Lian Li enables optimized thermal performance with PUE levels as low as 1.15. Every solution meets international certifications including RoHS, CE, and UL, supporting reliable and sustainable data center operations.

Meta Announces Landmark $10 Billion, 1-Gigawatt AI Data Center Expansion

Following the recent rollout of its commercial AI compute rental services, Meta is further accelerating its infrastructure roadmap. The company has officially launched a massive, vertically integrated computing expansion project to solidify the underlying foundation of its global artificial intelligence ecosystem.

Meta is committing $10 billion to construct a state-of-the-art supercomputing hub, marking its 33rd global data center. This hyper-scale facility features a monumental planned power capacity of 1 gigawatt (1,000 MW)—making it one of the largest single-campus data center developments in history, with a power footprint capable of matching 750,000 localized households.

Strategic Shift: Power-First Site Selection

This $10 billion capital injection represents complete vertical integration, encompassing land acquisition, structural construction, power infrastructure, advanced cooling systems, and next-generation IT hardware.

This initiative signals a fundamental shift in how hyper-scale operators approach infrastructure:

Grid Capacity Over Labor Markets: Traditional site selection prioritized proximity to fiber networks or large tech labor pools. Meta’s latest strategy prioritizes direct, high-capacity grid access and proximity to robust energy generation.

Aggressive Deployment Timelines: A $10 billion deployment underscores market urgency. Meta is moving faster than typical utility planning cycles, creating significant opportunities for independent power producers capable of delivering power capacity within a tight 18-to-24-month window.

Unlocking Deregulated Power Markets: Operators are increasingly favoring markets that offer direct market access, streamlined interconnection queues, and flexible power purchase agreements (PPAs)—such as the ERCOT system in West Texas and rapidly growing energy corridors in Western Canada—to sustain 1 GW workloads with reliable, dispatchable power.

Regional Economic and Infrastructure Impact

The flagship development will be located in Sturgeon County, Alberta, Canada, representing Meta’s first major data center investment in the region.

To guarantee an uninterrupted, resilient power supply without straining local resources, Meta is fully funding the development of brand-new grid infrastructure. The company is actively collaborating with a coalition of regional stakeholders, including Greenlight Limited Partnership, AltaLink, Capital Power, and the Alberta Electric System Operator (AESO), ensuring the facility is fully supported from day one.

Key Project Highlights:

Job Creation: The construction phase will create over 3,000 trade and construction jobs, followed by at least 300 long-term, full-time positions once the campus becomes fully operational.

Environmental Responsibility: Meta will fully absorb all utility and wastewater management fees associated with the facility’s operations, ensuring zero financial or environmental strain on the local municipality.

Silencing the "Compute Surplus" Debate

While Meta’s move to open up its idle capacity to the AI rental market initially sparked industry conversations regarding a potential "compute surplus," this $10 billion expansion confirms that Meta’s mid-to-long-term infrastructure scaling remains aggressive.

According to leading tech research firm SemiAnalysis, fears of overcapacity are unfounded. Meta’s data center construction and hardware procurement are projected to steadily accelerate, with capital expenditure (CapEx) expected to hit record highs by 2027.

Data shows that in the first half of 2026 alone, Meta secured over 5 gigawatts of compute capacity through third-party cloud and colocation agreements—a figure that does not even include its rapidly expanding self-built pipeline. Meta Founder and CEO Mark Zuckerberg noted that commercializing surplus capacity creates a sustainable, high-margin revenue stream, which in turn provides the stable cash flow needed to fund these massive, long-term infrastructure plays.

The Critical Role of Liquid Cooling and Lian Li’s Enterprise Solutions

As artificial intelligence workloads evolve from traditional LLM training to continuous, high-density inference, thermal management has become the primary bottleneck in data center design. To support 1 GW of high-density compute, Meta's new data center will deploy an advanced, water-saving closed-loop liquid-to-air dry cooling system, achieving zero-water consumption during standard cooling cycles.

In modern hyper-scale facilities, legacy air cooling is no longer sufficient to dissipate the extreme thermal design power (TDP) of next-generation AI accelerators. Liquid cooling has transitioned from an experimental alternative to an absolute operational necessity. By delivering liquid coolant directly to the chip, liquid cooling systems drastically reduce energy overhead, lower Power Usage Effectiveness (PUE) ratings, and allow data centers to pack more computing density into a smaller physical footprint.

As data centers worldwide race to retrofit for high-density AI, industry leaders are turning to specialized thermal architectures. Lian Li, globally recognized for its precision engineering and innovative thermal designs, provides cutting-edge liquid cooling components and custom enclosure solutions tailored to meet these demanding enterprise standards. From ultra-reliable industrial pump systems to highly efficient radiators and modular liquid-loop configurations, Lian Li’s thermal management expertise helps modern data centers maximize hardware efficiency, prevent thermal throttling, and achieve the rigorous sustainability goals required by the next generation of supercomputing hubs.

Can a Traditional Data Center Support AI Workloads? A Practical Guide

Can an existing data center support AI workloads? Explore the upgrades needed for GPU clusters, high rack density, liquid cooling, power capacity, storage, and networking.

Modular Liquid-Cooled Data Center Containers: Benefits, Design and Deployment

Explore modular liquid-cooled data center containers for AI, HPC, edge computing, and high-density workloads. Learn the benefits, technical components, deployment process, and supplier selection factors.

Read More:

CDU Cooling Systems Explained: Liquid-to-Air vs. Liquid-to-Liquid

An In-Rack Coolant Distribution Unit (CDU) is a compact, self-contained liquid cooling system mounted directly inside a server rack.

It isolates server liquid loops from the broader facility water supply, securely pumping coolant to chips and GPUs to support high-density AI and HPC workloads.

Key Advantages: Ultra-High Density: Supports computing densities of 30kW to 250kW+ per rack, which is virtually impossible with traditional data center air conditioning.

Space Saving: Mounts directly at the bottom or top of standard enclosures (often taking up just 4U to 6U of vertical space), preserving valuable data center floor space.

Risk Mitigation: Isolates potential leaks to a single rack, often featuring built-in redundant pumps, smart flow control, and leak-detection strings to protect mission-critical hardware.

Website：

NVIDIA Vera Rubin NVL72: Why Next-Generation AI Infrastructure Is Becoming a Multi-Million-Dollar Investment

The AI infrastructure race is entering a new phase. As model sizes continue to grow and inference workloads scale globally, the cost of deploying state-of-the-art AI systems is rising dramatically. NVIDIA's upcoming Vera Rubin NVL72 platform represents one of the most powerful AI computing architectures ever designed—but it also comes with an unprecedented price tag.

Industry estimates suggest that a fully configured Rubin NVL72 rack could cost between $7.8 million and $9 million, nearly doubling the investment required for previous-generation AI infrastructure. The increase is not driven solely by GPUs. Instead, memory, networking, storage, power delivery, and liquid cooling are becoming major contributors to overall system cost.

The Evolution of AI Rack Architecture

Unlike traditional GPU servers, the Rubin NVL72 is a rack-scale AI supercomputer. Each rack integrates:

72 NVIDIA Rubin GPUs

36 Vera CPUs

Next-generation NVLink switching fabric

High-speed networking infrastructure

Massive memory and storage subsystems

Advanced liquid cooling technologies

This tightly integrated architecture is designed to maximize AI training and inference performance while reducing latency across the entire system. NVIDIA has positioned Rubin as a platform capable of delivering substantially higher inference throughput and significantly lower cost per token than previous generations.

Memory Has Become the Largest Cost Driver

One of the most significant changes in the Rubin generation is the explosion of memory costs.

Recent industry analyses indicate that memory-related components—including HBM4, LPDDR5X, and high-capacity flash storage—now account for more than $2 million per rack, representing roughly one-quarter of total system cost. This marks a dramatic increase compared to earlier Blackwell-based platforms.

Several factors are contributing to this trend:

Growing AI model context windows

Increased demand for high-bandwidth memory

Larger inference cache requirements

Higher-capacity storage tiers for AI workloads

As a result, memory is no longer a supporting component—it has become a strategic resource that directly impacts infrastructure budgets.

Networking and System Integration Complexity

Modern AI clusters depend on far more than compute accelerators.

Rubin NVL72 introduces a highly interconnected architecture that requires advanced switching, networking controllers, and high-speed interconnect technologies. These components ensure that dozens of GPUs can operate as a unified system while maintaining extremely low communication latency.

The increasing complexity of rack-scale integration is pushing costs upward across:

NVLink fabrics

Ethernet and InfiniBand networking

PCB and packaging technologies

Power delivery systems

Rack-level orchestration hardware

This shift highlights a broader industry trend: AI infrastructure spending is moving from individual servers to complete system architectures.

Liquid Cooling Is No Longer Optional

Power density continues to rise with every AI generation.

Industry reports estimate that liquid cooling for a single NVL72-class rack can cost tens of thousands of dollars, with future Rubin systems expected to require even more advanced thermal management solutions. Specialized cold plates, coolant distribution units, and rack-level cooling infrastructure are becoming standard requirements for deployment.

As AI accelerators approach and exceed multi-kilowatt power envelopes, efficient cooling is increasingly viewed as a critical performance enabler rather than simply a facility requirement.

The Impact on Data Center Operators

The growing cost of AI infrastructure is reshaping data center investment strategies.

Large cloud providers and hyperscalers are expected to remain the primary buyers of Rubin-based systems because they can spread infrastructure costs across massive AI workloads. For enterprises and regional service providers, however, the financial barriers to deploying cutting-edge AI infrastructure are becoming increasingly difficult to overcome.

This trend is accelerating demand for:

AI colocation services

High-density liquid-cooled facilities

Infrastructure-as-a-Service platforms

AI-focused data center ecosystems

Organizations are increasingly evaluating whether to own AI infrastructure or consume it through cloud-based services.

What This Means for the Future of AI

The Vera Rubin generation demonstrates that the future of AI is not simply about faster GPUs. Success increasingly depends on balancing compute, memory, networking, power, and cooling within a single integrated platform.

As AI workloads continue to expand, the industry's focus is shifting from component performance to total system optimization. The companies that can efficiently deploy and operate these multi-million-dollar AI platforms will be best positioned to support the next wave of large-scale AI innovation.

Rubin NVL72 is more than a server rack—it is a preview of the infrastructure foundation that will power the future of artificial intelligence.

Lianli Liquid Cooling’s One-Stop Liquid Cooling Services for Data Centers:

As a leading provider of liquid cooling solutions for AI data centers, Lianli Liquid Cooling delivers products that reduce PUE to below 1.15 and achieve an overall energy-saving rate exceeding 30%. Its product portfolio—ranging from cold plates and liquid-cooled server racks to CDUs and containerized data centers—is deeply compatible with NVIDIA’s product series (such as the H100/H200 and GB200/GB300). These products hold authoritative international certifications, including RoHS, CE, and UL, and are available for rapid delivery.

To achieve continuous uptime in mission-critical data centers, modular architecture must be coupled with rigorous component selection. At Lian Li Liquid Cooling, we focus on engineering robust loop ecosystems:

🔹 Custom Cold Plates: Precision-machined for optimal thermal transfer at the chip level. 🔹 In-Rack & In-Row CDUs: Smart, redundant distribution systems that tightly regulate fluid temperature, flow rates, and pressure. 🔹 Liquid-Cooled Server Cabinets: Fully integrated enclosures designed for clean plumbing, leak prevention, and painless maintenance.

By utilizing top-tier, reliable manufacturing standards for every internal manifold, valve, and pump, we ensure your infrastructure remains operational 24/7/365.

Building a modular data center? Partner with an end-to-end manufacturer who owns the entire liquid loop architecture.

In 2026, the economics of data centers come down to two metrics: Power Usage Effectiveness (PUE) and Total Cost of Ownership (TCO). With AI server energy demands scaling exponentially, traditional air-chilled systems are becoming a financial and environmental liability.

Direct-to-chip liquid cooling saves up to 40% on facility energy costs compared to traditional air cooling. But how do colocation and hyperscale providers protect that capital investment as hardware rapidly evolves?

Universal Cold Plate Solutions.

By investing in direct liquid cooling architectures that are fully compatible across NVIDIA, AMD, and Intel form factors, operators realize massive strategic advantages:

1️⃣ Zero Stranded Capacity: Seamlessly swap out older server blades for next-gen AI accelerators without replacing the underlying fluid infrastructure. 2️⃣ Immediate PUE Reduction: Rejecting up to 80%+ of component waste heat directly into the liquid loop minimizes the need for massive, power-hungry CRAC units. 3️⃣ Sustainable Heat Reuse: Universal cold plates make it easier to capture high-temperature waste water, which can be routed directly into district heating or industrial processes.

Sustainability shouldn't mean sacrificing hardware flexibility. A universal cooling approach ensures your facility remains agile enough to host whatever hardware the market demands next, all while shrinking your carbon footprint.

Ready to de-risk your next infrastructure upgrade? Let’s connect.

Coolant Distribution Units (CDUs) in Liquid-Cooled Data Centers: A Technical Overview

Coolant Distribution Units serve as the central thermal management component in liquid-cooled data centers. Positioned between the facility’s chilled-water or dry-cooler plant and server cold plates, a CDU handles heat transfer, hydraulic isolation between loops, precise temperature regulation, and safety features such as leak detection and automatic shutdown. These units often represent 30–40% of the total liquid cooling system cost, making them critical for both performance and total cost of ownership. A failure can take entire racks offline, underscoring their importance.

Core Operating Principle

A typical CDU uses a brazed-plate heat exchanger to couple two separate fluid loops without mixing them.

Primary loop (facility side): Connects to building cooling infrastructure (chillers, cooling towers, or dry coolers). Typical parameters include supply temperatures of 7–12 °C (chilled water) or 30–40 °C (dry cooler), with returns 5–10 °C higher.

Secondary loop (IT side): Circulates deionized (DI) water or dielectric fluid through server cold plates, absorbing heat from CPUs and GPUs. Supply is usually 18–25 °C; return reaches 30–45 °C.

This isolation protects the ultra-pure secondary coolant from contamination while allowing independent pressure and temperature optimization. Variable-speed pumps, control valves, and a PLC with PID algorithms maintain stable operation by modulating flow and valve positions based on continuous sensor feedback.

System Architecture and Performance Metrics

Liquid cooling deployments follow a three-tier model:

Cooling source tier — Outdoor plants providing primary-loop capacity.

CDU distribution tier — Heat exchanger, pumps, valves, sensors, and controls.

IT load tier — Rack-level cold plates and manifolds.

Key thermal parameters vary by configuration: ParameterPrimary (Chilled)Primary (Dry Cooler)Secondary (IT)Supply Temp.7–12 °C30–40 °C18–25 °CReturn Temp.12–18 °C35–45 °C30–45 °CΔT5–6 °C5–10 °C10–20 °CPressure0.3–0.6 MPa0.2–0.5 MPa0.1–0.4 MPaCoolantWater/glycolGlycolDI water/dielectric

Typical CDU metrics include cooling capacities from 20 kW to over 2 MW, ≥95% heat exchange efficiency, ±0.5 °C temperature control accuracy, system PUE ≤1.10, and MTBF ≥80,000 hours.

Internal Components and Design

Core elements include:

Plate heat exchanger (e.g., 316L stainless, Alfa Laval, Tranter) for efficient transfer and isolation.

Variable-frequency pumps (Grundfos, Wilo, CNP) for secondary circulation.

Control valves, strainers, expansion vessels, sensors (PT100 RTDs, pressure transmitters, flow meters, conductivity probes, leak detection cables).

PLC + HMI for logic, PID control, and operator interface.

Piping connections combine flanged, grooved, TIG-welded, threaded, and quick-disconnect methods. Electrical labeling follows IEC standards. Control systems use a four-layer architecture (management/BMS, PLC/HMI, actuation, sensing) with Modbus/BACnet communications. Safety interlocks respond rapidly to leaks, overpressure, overtemperature, and fluid quality issues.

Form Factors and Quick-Connect Technology

Common designs include:

In-rack (2U–4U, 20–80 kW): Compact for edge/small deployments.

In-row/floor-standing (100–500+ kW): Balanced capacity and serviceability.

Modular/stackable: Scalable to multi-MW for hyperscale/AI clusters.

Quick-disconnect couplings (UQD, blind-mate, dry-break) from suppliers like Stäubli, CEJN, JONHON, and others minimize leakage (<0.05–0.1 mL) and support high-density blind mating. 316L stainless steel piping with proper welding, passivation, pressure testing, and flushing is standard. Copper is avoided in secondary loops to prevent conductivity rise.

Market Landscape, Challenges, and Trends

Leading vendors include Envicool, Vertiv, Huawei, ZTE, and others, with strengths in full-stack solutions, ecosystem integration, or modularity. Key challenges are leak risks, vendor lock-in from non-standard interfaces, pump reliability, water quality maintenance, and physical constraints (weight, access).

Mitigations and forward trends include:

Negative-pressure designs to prevent outward leaks.

Precise pump sizing with redundancy.

Continuous conductivity monitoring and ion-exchange.

AI/ML-enhanced controls for dynamic AI workloads.

Waste heat recovery from warmer return fluid.

Ongoing standardization via OCP, UQD, and national specifications to reduce lock-in and TCO.

Installation, Commissioning, and Maintenance

Site requirements cover floor loading, power, networking, drainage, and clearances. Construction emphasizes 316L piping, rigorous testing, flushing, and insulation. Commissioning involves filling/deaeration, PID tuning, interlock/redundancy verification, and a 72-hour burn-in. Acceptance focuses on pressure integrity, temperature stability, flow balance, leak response, and water quality.

Preventive maintenance includes daily monitoring, periodic strainer cleaning/sensor calibration, annual coolant replacement, and multi-year HEX servicing. Common troubleshooting addresses temperature/pressure deviations, leaks, conductivity rise, and communications issues.

Conclusion

As AI-driven rack densities climb, CDUs have become essential infrastructure for efficient, reliable liquid cooling. Success depends on robust design, proper component selection, standardization, meticulous installation, and disciplined operations—particularly around leak prevention, fluid quality, and controls. With advancing modularity, negative-pressure tech, and interoperability standards, liquid cooling is transitioning from specialized deployments to a mainstream, cost-effective solution for high-performance data centers.

Cryptocurrency has evolved from a speculative niche into an increasingly mature asset class attracting retail investors, institutions, and global enterprises. While short-term price swings still dominate headlines, long-term investors are increasingly focusing on projects with strong fundamentals, real-world adoption, sustainable ecosystems, and resilience across market cycles.

If you're building a long-term crypto portfolio for 2026 and beyond, diversification matters. Rather than chasing trends, investors often look for assets that combine market leadership, technological innovation, infrastructure value, and risk management.

This guide explores six cryptocurrencies frequently considered for long-term investment potential: Bitcoin (BTC), Ethereum (ETH), Solana (SOL), Polygon (POL), Chainlink (LINK), and PAX Gold (PAXG).

1. Bitcoin (BTC) — The Digital Store of Value

Why Bitcoin remains a long-term favorite

Bitcoin continues to hold a unique position as the first and most recognized cryptocurrency. Often referred to as “digital gold,” Bitcoin’s value proposition centers on scarcity, decentralization, and security.

With a fixed supply cap of 21 million coins, Bitcoin is designed to resist inflation over the long term. Institutional participation, expanding custody infrastructure, and broader regulatory clarity in major markets have strengthened its position as a strategic portfolio asset.

Long-term strengths:

Limited supply creates scarcity

Highest market recognition and liquidity

Increasing institutional adoption

Strong network security and decentralization

Potential risks:

Lower growth multiples compared to smaller cryptocurrencies

Regulatory changes may affect adoption speed

Limited programmability compared with smart contract platforms

Best suited for: Conservative crypto investors seeking long-term exposure with relatively lower ecosystem risk.

2. Ethereum (ETH) — The Foundation of Web3

Why Ethereum still matters in 2026

Ethereum remains the leading smart contract ecosystem powering decentralized finance (DeFi), NFTs, tokenization, and thousands of blockchain applications.

Unlike Bitcoin’s primary role as digital value storage, Ethereum functions as programmable infrastructure. Developers use Ethereum to build applications ranging from financial services to enterprise blockchain solutions.

As scaling technologies continue improving transaction efficiency, Ethereum’s long-term investment thesis increasingly centers on network utility and ecosystem dominance.

Long-term strengths:

Largest developer ecosystem

Dominant smart contract platform

Broad institutional interest

Continuous protocol evolution

Potential risks:

Competition from faster chains

Network congestion during peak demand

Regulatory uncertainty around staking models

Best suited for: Investors seeking exposure to blockchain infrastructure and long-term ecosystem growth.

3. Solana (SOL) — High-Speed Blockchain for Scale

Why Solana attracts growth investors

Solana has positioned itself as one of the leading high-performance blockchain networks by focusing on speed, low transaction costs, and user experience.

Its ecosystem supports decentralized applications, gaming platforms, payment solutions, and emerging consumer-facing blockchain products.

Long-term investors often view Solana as a higher-growth, higher-volatility complement to Bitcoin and Ethereum.

Long-term strengths:

Fast transaction throughput

Low operating costs

Growing developer activity

Strong appeal for consumer applications

Potential risks:

Greater volatility

Competition from multiple Layer-1 networks

Historical concerns around network reliability

Best suited for: Investors comfortable with higher risk in exchange for higher growth potential.

4. Polygon (POL) — Scaling Blockchain Adoption

Why Polygon deserves attention

Polygon focuses on improving blockchain scalability and reducing transaction costs for decentralized applications.

Rather than competing directly with Ethereum, Polygon has built much of its strategy around helping Ethereum scale efficiently. This positioning gives Polygon exposure to broader blockchain adoption without relying solely on independent network growth.

Its ecosystem includes partnerships across gaming, enterprise initiatives, and Web3 infrastructure.

Long-term strengths:

Strong compatibility with Ethereum

Lower transaction costs

Growing enterprise engagement

Expanding Layer-2 ecosystem

Potential risks:

Dependence on Ethereum ecosystem growth

Intense competition among scaling solutions

Best suited for: Investors looking for infrastructure exposure with moderate growth potential.

5. Chainlink (LINK) — Connecting Blockchain to Real Data

Why Chainlink stands out

Smart contracts are powerful, but they require reliable external data to function effectively. Chainlink provides decentralized oracle services that connect blockchain applications with real-world information.

From financial market data to insurance triggers and tokenized assets, Chainlink plays a critical infrastructure role across the crypto ecosystem.

Many long-term investors view Chainlink as a “picks-and-shovels” investment—supporting the entire blockchain economy rather than competing with other chains.

Long-term strengths:

Critical infrastructure role

Broad integration across blockchains

Growing utility for tokenized assets

Strong developer adoption

Potential risks:

Complex valuation compared with traditional assets

Adoption depends partly on broader blockchain growth

Best suited for: Investors seeking infrastructure exposure beyond Layer-1 networks.

6. PAX Gold (PAXG) — Digital Gold with Stability

Why PAX Gold may improve portfolio balance

PAX Gold takes a different approach from traditional cryptocurrencies. Each token is designed to represent ownership tied to physical gold reserves.

For investors who believe in blockchain technology but want reduced volatility, PAX Gold offers an alternative that combines digital accessibility with exposure to a historically defensive asset.

It may serve as a stabilizing allocation alongside growth-oriented crypto holdings.

Long-term strengths:

Backed by physical gold exposure

Lower volatility compared with many cryptocurrencies

Useful diversification tool

Potential risks:

Limited upside compared with growth assets

Performance tied to gold market dynamics

Best suited for: Investors seeking portfolio stability and diversification.

Sample Long-Term Crypto Allocation

A balanced approach could look like:

Bitcoin: 35–45%

Ethereum: 25–35%

Solana: 10–15%

Polygon: 5–10%

Chainlink: 5–10%

PAX Gold: 5–15%

Allocation should reflect your risk tolerance, investment horizon, and liquidity needs.

Power Smarter Cryptocurrency Mining with Advanced Liquid Cooling

Cryptocurrency continues to attract long-term investors worldwide, and mining remains one of the most strategic ways to participate in the digital asset economy. Whether mining Bitcoin, Dogecoin, or other proof-of-work cryptocurrencies, maintaining stable hardware performance is essential for maximizing output and controlling operating costs.

Modern mining environments place enormous demands on ASIC equipment. Excess heat can reduce efficiency, shorten hardware lifespan, and increase energy consumption. That’s why advanced cooling technology has become a critical part of successful mining infrastructure.

Lian Li offers high-performance liquid cooling solutions purpose-built for cryptocurrency mining applications. Designed to support mainstream mining equipment, Lian Li systems provide flexible deployment options including direct liquid cooling and immersion cooling technologies to meet the requirements of different mining scales.

By improving heat dissipation and thermal management, Lian Li helps mining operators enhance operational stability, optimize energy usage, and maintain consistent performance under continuous workloads.

Every solution is developed to meet internationally recognized RoHS, CE, and UL standards, delivering dependable quality and operational safety. Backed by extensive industry expertise, Lian Li has successfully deployed more than 5 million liquid cooling units, contributing to an installed cooling capacity exceeding 10 GW across global projects.

As cryptocurrency mining continues to evolve, efficient infrastructure is becoming a competitive advantage. With advanced cooling technology and proven deployment experience, Lian Li helps build reliable, scalable, and future-ready mining operations.

Final Thoughts

Long-term cryptocurrency investing is rarely about finding the next overnight winner. More often, success comes from selecting assets with durable use cases, maintaining diversification, and staying invested through market cycles.

Bitcoin and Ethereum remain foundational holdings for many investors. Solana and Polygon represent growth and infrastructure opportunities. Chainlink supports the broader blockchain economy, while PAX Gold can add a defensive layer to a crypto portfolio.

As always, conduct independent research, understand the risks, and avoid investing more than you can afford to hold through volatility.