The rapid expansion of artificial intelligence, cloud computing, hyperscale infrastructure, and high-density computing is forcing a fundamental redesign of how data centers are monitored and controlled. Unitronics’ technical paper, “PLCs Profit from the Data Center Automation Revolution,” examines how programmable logic controllers (PLCs) integrated with the industry-standard Redfish management protocol can simplify data center automation while improving reliability, energy efficiency, and operational control. Engineers, facility managers, automation professionals, and data center developers interested in the original publication can find it on the Unitronics website:  PLCs Profit from the Data Center Automation Revolution⁠. This article evaluates the technical concepts presented in that paper, explains their significance, and places them within the broader context of modern data center infrastructure.  

Rather than focusing on servers themselves, the white paper addresses one of the fastest-growing engineering challenges facing modern infrastructure: how to automate the physical systems that allow AI and cloud computing facilities to operate safely, efficiently, and continuously. It argues that PLC technology—long proven in industrial automation—can become a core component of next-generation data center infrastructure when integrated directly into modern IT management architectures through native Redfish support.  

The document is written primarily for:

* Automation engineers

* Controls engineers

* Mechanical engineers

* Data center designers

* Facility managers

* Infrastructure developers

* Cooling system manufacturers

* System integrators

* Operations managers

* Industrial PLC developers

Unlike many marketing documents that concentrate on server hardware, networking, or AI processors, this paper concentrates on the physical infrastructure responsible for keeping those processors operating reliably.

The paper begins by documenting the extraordinary growth expected in worldwide data center investment. It cites industry forecasts projecting growth from approximately $386 billion in 2025 to more than $1 trillion by 2034, while identifying AI, cloud computing, Internet of Things deployments, remote services, and exploding digital data generation as the primary drivers. It further notes that automation systems themselves represent a rapidly expanding market because every new data center requires increasingly sophisticated environmental monitoring and facility control systems.  

From a technical perspective, the growth projections themselves are reasonable. Although different research firms produce somewhat different market estimates, nearly every major analyst forecasts sustained double-digit expansion in AI infrastructure over the coming decade. The exact dollar values should be viewed as representative forecasts rather than precise predictions, but the overall market trajectory is well supported.

The paper then transitions to programmable logic controllers.

PLCs have been the foundation of industrial automation for decades because they provide deterministic real-time control. Unlike general-purpose computers that run many competing software tasks, PLCs continuously execute predefined control logic with predictable timing. This makes them ideal for controlling equipment such as:

* Pumps

* Cooling towers

* Chillers

* Heat exchangers

* Variable-frequency drives

* Fan systems

* Valve networks

* Water treatment equipment

* Emergency shutdown systems

These are precisely the systems that determine whether a high-density AI data center remains operational.

One particularly important point made in the paper concerns cooling.

Modern AI clusters frequently consume tens of kilowatts per rack, with leading GPU systems now exceeding 100 kW per rack in some installations. Traditional room-air cooling becomes increasingly inefficient at these power densities, forcing the industry toward liquid cooling technologies.

The white paper discusses several major cooling architectures:

* Traditional CRAC and CRAH air cooling

* Direct-to-chip liquid cooling

* Liquid immersion cooling

* Hybrid cooling systems

These represent the major commercial cooling approaches currently deployed throughout the industry. The discussion is technically accurate, although necessarily brief.  

The document correctly explains that sophisticated automation is becoming increasingly important because cooling systems must dynamically respond to rapidly changing computational loads. AI training jobs can create dramatic thermal changes across hundreds or thousands of servers within minutes.

Real-time PLC control allows cooling systems to adjust:

* Pump speeds

* Fan speeds

* Flow rates

* Valve positions

* Cooling loop temperatures

* Equipment sequencing

This reduces energy consumption while maintaining safe operating temperatures.

 The paper cites an estimate that approximately 40% of data center energy consumption is devoted to cooling. That figure is representative of many legacy facilities but should not be interpreted as universally applicable. Modern hyperscale facilities employing advanced liquid cooling, economizers, free cooling, or warm-water cooling often achieve significantly lower cooling energy percentages. Nevertheless, cooling remains one of the largest operational energy consumers, making optimization economically important.  

Perhaps the strongest technical section discusses communication protocols.

Traditional industrial PLCs communicate using protocols such as:

* Modbus

* EtherNet/IP

* PROFIBUS

* DeviceNet

Meanwhile, modern servers, storage systems, and infrastructure management platforms increasingly communicate using Redfish.

Historically these worlds required protocol translators or gateways.

Every gateway introduces:

* Additional latency

* Additional hardware

* Additional software

* Additional maintenance

* Additional cybersecurity exposure

* Additional failure points

The paper argues that embedding native Redfish support directly into PLCs removes much of this integration complexity. Technically, this is a significant architectural improvement because eliminating protocol conversion simplifies system design while improving interoperability.  

The discussion of Redfish deserves special attention.

Redfish is an open standard developed by the Distributed Management Task Force (DMTF) to replace older server-management protocols such as IPMI. It uses modern RESTful APIs together with JSON data structures, making it compatible with contemporary software development tools, orchestration systems, and cloud management platforms.

Instead of relying on proprietary interfaces, Redfish provides standardized methods for monitoring and controlling:

* Servers

* Storage

* Networking

* Power systems

* Thermal equipment

* Environmental sensors

The paper correctly identifies Redfish as a major step toward interoperability across multi-vendor data center environments.  

One of the document’s strongest discussions involves thermal equipment schemas.

Recent Redfish specifications include standardized object models for:

* Coolant distribution units

* Pump assemblies

* Heat exchangers

* Reservoirs

* Leak detection

* Rack cooling

* Liquid cooling loops

* Facility cooling infrastructure

Standardized schemas significantly reduce integration effort because software developers no longer need vendor-specific interfaces for every component.

The document also discusses direct sensor integration.

Instead of merely reporting equipment status, Redfish-enabled PLCs can continuously collect environmental information including:

* Temperature

* Humidity

* Airflow

* Equipment state

* Pump status

* Cooling loop conditions

This information can then support predictive maintenance, AI-assisted facility optimization, automated fault detection, and improved operational awareness.  

The competitive positioning section requires additional context.

The paper states that Unitronics is currently the only manufacturer offering PLCs with native Redfish support. That may have been accurate when the paper was published, but it should be treated as a time-sensitive product claim rather than a permanent industry fact. Major automation suppliers—including Siemens, Schneider Electric, Rockwell Automation, Beckhoff, ABB, Phoenix Contact, and others—continue to expand their integration capabilities for modern IT environments. Organizations evaluating automation platforms should verify current product capabilities before making procurement decisions.  

Compared with conventional industrial automation approaches, native Redfish integration offers several practical advantages:

* Reduced engineering complexity

* Fewer gateway devices

* Lower maintenance requirements

* Better interoperability

* Faster deployment

* Simplified cybersecurity management

* Improved compatibility with Data Center Infrastructure Management (DCIM) platforms

* Easier integration with AI-driven facility management systems

These advantages become increasingly valuable as facilities grow from tens of megawatts to hundreds of megawatts and eventually to gigawatt-scale campuses.

The paper’s greatest strength is that it recognizes an important industry shift. Historically, industrial operational technology (OT) and enterprise information technology (IT) evolved independently, using different hardware, software, communication standards, and engineering disciplines. Modern AI data centers are forcing these worlds together. Cooling plants, electrical systems, backup power, liquid distribution networks, and facility controls must increasingly exchange real-time information with servers, orchestration software, and infrastructure management platforms. Standards such as Redfish provide a practical framework for that convergence.

Although the publication serves as a product-oriented technical paper and naturally emphasizes Unitronics’ own technology, its discussion of PLCs, Redfish, cooling automation, and interoperability is technically sound and reflects broader trends across the data center industry. As artificial intelligence, cloud computing, scientific research, advanced manufacturing, and digital services continue driving unprecedented demand for computing capacity, the supporting physical infrastructure becomes every bit as important as the processors performing the computations. Reliable automation, standardized communications, intelligent cooling control, and seamless integration between operational technology and information technology will determine how efficiently the next generation of hyperscale facilities can be designed, constructed, and operated. Technologies that simplify interoperability while improving reliability and energy efficiency will play an increasingly important role in delivering the resilient, scalable, and sustainable computing infrastructure the United States will require to remain globally competitive in the decades ahead.