Evolving Power Management

As facility managers look to support evolving data center power needs, they must take a strategic approach to power management – systems that support both IT and OT – to prevent disruptions and meet the growing demands of AI-supporting operations. Recent and emerging advancements in both physical and digital power management technology can help shape this approach.

Build the right foundation

In the data center, effective power protection starts in the grey space, where advanced solutions can help maintain resilience as infrastructure becomes more complex and power-intensive. Examples include:

• Uninterruptible power supplies (UPSs) – UPSs, which provide critical battery backup power, are the centerpiece of data center resiliency and serve as a bridge to generator power in the event of an outage. As infrastructure has expanded to accommodate AI workloads, more power-intensive graphical processing units (GPUs) and high-density storage, UPS technology has evolved to offer:

  • Higher efficiency and power density in a smaller footprint

  • Flexible battery technology, including lithium-ion options for extended lifespan and reduced maintenance.

  • Faster deployment options to meet rapid scalability needs.

  • Easier maintenance

For data centers with aging infrastructure, some manufacturers offer UPS modernization programs that retrofit existing frames with updated electronics and batteries, allowing for faster, cost-effective upgrades without requiring a full system replacement.

• Medium-voltage switchgear – Medium-voltage switchgear provides centralized control and protection for power equipment, including generators, motors, feeder circuits and transmission lines. Modern switchgear enhances safety by protecting personnel from arc faults on both grounded and ungrounded systems. FMs should prioritize solutions that provide:

  • maintenance-free medium-voltage compartments.

  • compact footprints to optimize space.

  • compliance with IEEE C37.20.9 and IEC 62271-200 safety standards.

  • remote draw-out capabilities.

  • ohmic voltage sensing (OVS), which increases reliability over traditional core and coil voltage transformers while saving space.

Some modern switchgear can also adapt to different voltages without requiring replacement, offering greater flexibility in dynamic energy environments.

• Molded case circuit breakers – Advancements in circuit breaker technology can help improve protection against arc flash events while enabling smaller, more adaptable designs. Key features of these next-generation circuit breakers include:

  • electronic trip units for real-time monitoring of breaker health and thermal status.

  • embedded communications and metering to optimize power distribution and reduce costs.

  • predictive diagnostics to identify and address potential failures before they occur.

• Floor power distribution – Floor-standing and cabinet power distribution units (PDUs) manage incoming power and distribute it to individual or grouped server racks. Floor-standing and cabinet PDUs distribute power to server racks while ensuring:

  • load balancing and voltage regulation.

  • harmonic reduction to improve power quality.

  • monitoring capabilities to detect potential threats and anomalies.

As data centers scale, modern PDUs offer greater configurability and flexibility, allowing engineers to customize solutions based on their specific power distribution needs. Some manufacturers now offer the opportunity to retrofit existing PDUs with modern equipment inside an existing PDU frame, delaying the need to fully replace current PDUs as requirements change.

• Battery energy storage – Containerized battery energy storage systems provide backup power, reduce peak energy costs and enable participation in grid energy sales. These solutions can also support renewable energy integration by storing locally produced power for building operations or electric vehicle charging.

Protect the production floor

While not typically the domain of FMs, the white space – where active servers and computing equipment are housed – is helping drive the evolution of data centers in the AI era. It is essential that FMs become familiar with advancements in these technologies, as the convergence of IT and OT systems requires a more complete understanding of all systems required to maintain uptime. FMs should become familiar with:

• IT rack enclosures – Modern IT racks are designed to organize and secure equipment in high-density data centers and edge environments. Newer enclosures, especially those designed using Open Compute standards, can support heavier AI servers and critical IT infrastructure, offering higher weight capacities, deeper enclosures and different designs for better visibility in high-density environments. Many modern racks are also being designed to accommodate alternative cooling methods, such as liquid cooling and chimney solutions that can remove heat without infiltrating the cold air supply.

• Aisle containment systems – Aisle containment systems are a simple way to improve cooling efficiency in hot aisle/cold aisle rack configurations. Essentially creating a room within the aisle, these systems help keep hot and cold air separated to make existing air conditioning systems more efficient. Modern advancements in aisle containment include chimney systems that remove heat without infiltrating the cold air supply, a needed innovation as AI computing equipment becomes hotter.

• Liquid cooling – The amount of heat produced by data center infrastructure is rising as operators invest in higher density systems to support AI. Traditional, fan-driven cooling is insufficient to meet the demands of these new technologies. Liquid cooling is a relatively new approach that improves heat transfer using dielectric fluids. Two types of liquid cooling in particular show promise:

  • Direct-to-chip cooling uses specialized liquid to extract heat at the source (i.e., directly in contact with CPUs and GPUs).

  • Immersion cooling, which requires full submersion of the server in nonconducting, dielectric fluids to improve overall cooling efficiency.

Both options offer distinct advantages and will be imperative as data centers experience higher power requirements through the proliferation of GPU servers. Because liquid cooling is nascent in its evolution and many data centers may not be equipped to adopt certain approaches, FMs may determine that a hybrid approach – using air and liquid cooling – is best to meet current needs while they explore new innovations.

• Rack power distribution – Rack power distribution units – devices that distribute power to IT equipment within a rack – have evolved with new capabilities to save time, reduce costs and simplify power management. Examples of recent innovations in rack PDU technology include:

  • higher power densities, fitting more outlets into the same space.

  • universal input compatibility, allowing a single SKU to support global deployments.

  • remote monitoring and management via network connectivity.

  • universal output compatibility: outlets that can accept both C13 and C19 plugs.

• Track busway – Track busway systems offer a structured, flexible alternative to traditional cable and conduit for safe and reliable power distribution. Modern track busway systems provide:

  • high-speed monitoring.

  • key electrical performance metrics.

  • proactive management of power usage and availability.

Additionally, these systems have evolved to offer considerable flexibility and configurability, both essential considerations in modernizing and future-proofing data centers to support emerging innovations. For example, modern track busway systems can include a multitude of features, such as tap-off boxes that are easily adapted to different layouts, with a multirange ampacity bus to support a wide range of power loads.

Protect energy availability & the grid

As data centers expand computing infrastructure to meet rising power requirements, many now realize that the existing energy grid may not sustain their growing needs. Data centers must also prepare for more frequent outages and disturbances as grid reliability may be compromised by escalating demand and an increasing reliance on renewable energy sources.

Deploying UPSs and battery energy storage systems with lithium-ion battery technology can help extend service life and offer higher power density in a smaller footprint – reserving room for new infrastructure to better support AI. An additional advantage of lithium-ion batteries, especially as sustainability demands rise, is their inherent energy storage capabilities. When combined with intelligent energy management, data analytics and sophisticated controls, a lithium-ion battery-powered UPS can be transformed into a distributed energy resource (DER), enabling a bidirectional flow of energy to and from the grid. This approach allows data centers to more efficiently and effectively coordinate multiple energy sources, anticipate energy needs and participate in grid programs to earn new revenue streams. Utilizing this approach can help curtail peak power draw from the utility, reducing costly demand charges and helping advance sustainability goals.

Deploying a full-scale microgrid is another promising strategy to enhance energy resilience, reduce grid dependence, advance islanding capabilities and optimize energy usage. Microgrids offer reliable backup power during unexpected outages by balancing fluctuations in energy demand and consumption. By leveraging microgrid technology, data centers can better support the integration of renewable energy sources like solar and battery energy storage, further reducing energy costs and boosting resiliency.

Protect power devices from cyber threats

While power management has only recently moved to the forefront of cybersecurity concerns, the integration of network connectivity with power management is changing the landscape. Devices like UPSs and PDUs are now frequently a part of network infrastructure and edge applications as data center operators seek to harness capabilities such as remote monitoring and management.

Major cyberattacks have highlighted vulnerabilities in critical infrastructure this year, and data centers will continue to be prime targets. As data center facilities plan for their cybersecurity priorities, it is important to prioritize protecting connected power devices from potential threats.

Global safety standards organizations, including Underwriters Laboratories (UL) and the International Electrotechnical Commission (IEC), provide FMs with important guidelines for the implementation of appropriate cybersecurity safeguards in network-connected devices. Deploying UPSs with network management cards that carry UL 2900-1 and ISA/IEC 62443-4-2 certifications in distributed environments, such as network closets, can provide peace-of-mind that devices have built-in cybersecurity features and capabilities to better protect infrastructure against breaches.

As organizations manage a wider range of connected UPSs over a distributed network in their facilities, many will seek to execute remote firmware updates to keep their devices up to date with the latest features. Because of this, FMs should look for power devices that require cryptographic signature checks to update firmware and avoid cybersecurity risks. FMs should seek vendors that offer 24/7 monitoring across converged IT/OT environments, as this can help add an extra layer of protection and visibility for critical infrastructure.

It is also important to follow cybersecurity best practices for power devices in distributed environments, just as with any other connected components. This includes:

• using firewall and industrial security solutions and encrypting information.

• hardening devices to only be able to operate as intended and conduct routine security assessments.

• regularly updating software and firmware (including antivirus definitions and antispyware).

• using advanced email filtering and establishing strong authentication and password policies with role-based controls.

• deploying endpoint protection and integrating employee cybersecurity awareness training.

Protect investments in mission critical equipment

As important as it is that FMs are strategic about their investments in mission-critical equipment, it is equally imperative that they take appropriate measures to protect those investments and ensure uninterrupted operations as the cost of infrastructure increases. Many vendors offer value-added services that can help extend the life of equipment and even upgrade without ripping and replacing critical systems. Operators should explore all available opportunities to leverage services from providers that can add an extra layer of protection, including:

• service-level agreements that cover an extended period and can be tailored to the facility’s unique needs and budget.

• preventive maintenance, which can include replacing equipment, such as UPS batteries, to help avoid downtime and other serious problems before they occur.

• remote monitoring services that provide 24/7 monitoring of vital systems and proactively alert the FM if equipment is experiencing issues; some of these services can also include predictive analytics, which leverage system data to predict the precise timing of potential issues.

Protect the future

As data center infrastructure evolves to support AI and high-performance computing, power management technology that serves both the grey space and white space will evolve in parallel. FMs educated about these advancements will be better equipped to protect their facilities while increasing efficiency, resilience and security. By adopting a strategic approach to power management, FMs can drive sustainable growth while securing their infrastructure for the next wave of innovation.