{"id":3720,"date":"2026-03-13T16:09:25","date_gmt":"2026-03-13T16:09:25","guid":{"rendered":"https:\/\/www.apexwaves.com\/blog\/?p=3720"},"modified":"2026-03-13T16:09:26","modified_gmt":"2026-03-13T16:09:26","slug":"inside-nis-approach-to-modern-data-center-power-monitoring","status":"publish","type":"post","link":"https:\/\/www.apexwaves.com\/blog\/inside-nis-approach-to-modern-data-center-power-monitoring\/","title":{"rendered":"Inside NI\u2019s Approach to Modern Data Center Power Monitoring"},"content":{"rendered":"\n

The impact of Artificial Intelligence (AI)<\/strong> in our day-to-day lives goes far beyond convenient chatbots. The rapid development of these AI tools and services has required the deployment of high-performance servers, resulting in a high-power density for data centers. As power needs increase, so does the need for reliable power monitoring. This post examines NI\u2019s powerful approach to data center power monitoring, which utilizes modular hardware and software to achieve the same vital measurements as conventional techniques, with added flexibility and scalability.<\/p>\n\n\n\n\n\n\n\n

How Data Centers Operate<\/strong><\/h2>\n\n\n\n

A data center<\/a><\/strong> is a physical location where networked computers collaborate to process, store, and distribute data. Companies rely on these centers for the delivery of online services, and they require substantial infrastructure to support the necessary hardware and software. The infrastructure needed to power these data centers is far more complicated than that of a normal home!\u00a0<\/p>\n\n\n\n

Inside a typical data center, you will find systems for cooling, ventilation, and fire suppression. You will also find uninterruptible power supplies, backup generators, and power subsystems. Links to external networks are another vital part of the architecture. <\/p>\n\n\n\n

\"Female<\/figure>\n\n\n\n

Data centers use a lot of energy.<\/strong> AI models have introduced a whole new level of computational complexity, and as digital transformation picks up speed, power needs are increasing. For example, AI chatbots often cannot function without a constant and stable network connection to their servers. These needs have led to the rise of AI data centers<\/strong>, which differ from their more traditional counterparts by prioritizing parallel processing and streamlined workflows for training and implementing machine learning models. These AI-specific data centers often include integrated security, high-performance graphics processing units (GPUs), tensor processing units (TPUs), and software-defined networking.\u00a0<\/p>\n\n\n\n

The Importance of Power Monitoring<\/strong><\/h2>\n\n\n\n

In a report<\/a>, the International Energy Agency <\/strong>posited that data center electricity consumption was projected to reach 415 terawatt hours (TWh), or over 1.5% of the world’s total electricity consumption, in 2024. They cited that over the past five years, this electricity consumption has increased annually at a rate of 12%.<\/p>\n\n\n\n

As this report makes clear, power is quickly becoming the largest operational cost for many data centers. Reducing energy usage is not only good for lowering costs for data center operators\u2014it also benefits the environment. Environmental impact is an important topic of discussion when it comes to data centers. Through effective power measurement systems, data centers can adhere to international energy and environmental guidelines while still meeting growing demands.<\/p>\n\n\n\n

Careful power monitoring is also crucial for maintaining equipment, boosting uptime, and optimizing energy. Data center operators want to provide their clients with dependable service, so uptime and maintenance go hand in hand. Monitoring is also the first line of defense for preventing outages, detecting anomalies, and maintaining efficiency.\u00a0<\/p>\n\n\n\n

\"Closeup<\/figure>\n\n\n\n

The \u201cBlack Box\u201d Problem<\/strong><\/p>\n\n\n\n

The major power monitoring challenge that NI\u2019s approach addresses is that the rapid growth of AI makes it hard to estimate energy usage. In fact, these workloads are often called \u201cblack boxes.\u201d \u201cBlack box\u201d AI systems<\/a><\/strong> describe situations in which the inputs and outputs are visible to the user, but the exact mechanics and logic that the AI tool uses to generate those outputs cannot be observed and are unknown. As a result, AI energy is incredibly difficult to characterize.<\/p>\n\n\n\n

NI\u2019s solution needs to both measure and adapt to the changing needs of AI. Post-processing is no longer a viable option for these applications that need actionable data\u2014real-time monitoring is a must.<\/p>\n\n\n\n

NI\u2019s Modular Approach to Power Monitoring<\/strong><\/strong><\/h2>\n\n\n\n

NI has long been a trusted manufacturer of easily accessible and dependable test and measurement products that are relied on for gathering crucial data across the globe.<\/p>\n\n\n\n

In a recent application note, NI focused on the CompactRIO platform<\/strong> as an example configuration of their methodology. This diverse and modular platform offers a tightly integrated software toolchain, high-performance processing capabilities, and I\/O that has been conditioned to specific sensors. Equipped with real-time processing, the cRIO platform is an excellent choice for applications requiring monitoring or control.\u00a0<\/p>\n\n\n\n

\n
\n
\"Angled<\/figure>\n<\/div>\n\n\n\n
\n
\"Angled<\/figure>\n<\/div>\n<\/div>\n\n\n\n

Modular instrumentation offers the ability to scale monitoring systems, customize the measurement configurations, and integrate with existing infrastructure. The core hardware components that NI highlighted for the example included the cRIO-9040<\/a><\/strong>, the NI-9244<\/a><\/strong>, and the NI-9238<\/a><\/strong>.\u00a0<\/p>\n\n\n\n

The cRIO-9040 embedded CompactRIO controller<\/strong> is equipped with a real-time processor and 4 slots, making it an excellent choice for enhancing the capabilities of the C Series modules. The NI-9244 voltage input module<\/strong> was used to measure each phase\u2019s voltage directly, making use of its 3 analog inputs and screw terminal connection. Finally, the NI-9238 voltage input module <\/strong>was connected to current sensors to accurately measure each phase\u2019s current.\u00a0<\/p>\n\n\n\n

Software was another important consideration for the example configuration. The LabVIEW Real-Time Module and the LabVIEW Electrical Power Toolkit were used to develop applications on the cRIO\u2019s Linux Real-Time OS platform and to provide the IP for performing power analysis.<\/p>\n\n\n\n

With a demonstration code, NI showed that the aforementioned modular hardware setup was more than capable of meeting the power monitoring requirements. This code showcased the software\u2019s ability to filter for specific frequency ranges, show an overview of system calculations, show raw waveforms, and calculate active power at various rates. <\/p>\n\n\n\n

THE FUTURE OF FLEXIBLE POWER MONITORING<\/strong><\/h3>\n\n\n\n

While AI workloads continue to evolve and create high energy demands, data centers can utilize NI\u2019s method to improve their operational performance and resilience. This flexible approach to power monitoring will shift the current landscape of reactive diagnostics into more predictive analytics. <\/p>\n\n\n\n

Going into the future, modular hardware and software such as the CompactRIO platform can be used together to achieve more effective and dependable power monitoring for data centers.<\/p>\n\n\n\n

ADDITIONAL INFORMATION<\/strong><\/p>\n\n\n\n

\n
NI Application Note<\/a><\/div>\n\n\n\n
IEA Special Report<\/a><\/div>\n<\/div>\n\n\n\n
<\/div>\n\n\n\n

<\/p>\n","protected":false},"excerpt":{"rendered":"

The impact of Artificial Intelligence (AI) in our day-to-day lives goes far beyond convenient chatbots. The rapid development of these AI tools and services has required the deployment of high-performance servers, resulting in a high-power density for data centers. As power needs increase, so does the need for reliable power monitoring. This post examines NI\u2019s […]<\/p>\n","protected":false},"author":1,"featured_media":3721,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[8,10],"tags":[139,1018,318,377,393,437,1019],"class_list":["post-3720","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-national-instruments","category-technology","tag-compactrio","tag-data-centers","tag-labview","tag-modular-systems","tag-national-instruments","tag-ni","tag-power-monitoring"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.apexwaves.com\/blog\/wp-json\/wp\/v2\/posts\/3720","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.apexwaves.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.apexwaves.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.apexwaves.com\/blog\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.apexwaves.com\/blog\/wp-json\/wp\/v2\/comments?post=3720"}],"version-history":[{"count":1,"href":"https:\/\/www.apexwaves.com\/blog\/wp-json\/wp\/v2\/posts\/3720\/revisions"}],"predecessor-version":[{"id":3728,"href":"https:\/\/www.apexwaves.com\/blog\/wp-json\/wp\/v2\/posts\/3720\/revisions\/3728"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.apexwaves.com\/blog\/wp-json\/wp\/v2\/media\/3721"}],"wp:attachment":[{"href":"https:\/\/www.apexwaves.com\/blog\/wp-json\/wp\/v2\/media?parent=3720"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.apexwaves.com\/blog\/wp-json\/wp\/v2\/categories?post=3720"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.apexwaves.com\/blog\/wp-json\/wp\/v2\/tags?post=3720"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}