If you have ever tried to improve sustainability without good measurements, you know the feeling. It is like trying to tune a database with your eyes closed while the workload keeps changing and someone keeps flipping breakers in the data center. Fun for no one. The good news is that we finally have first class power visibility where it matters most. Inside the platform that runs your apps. At the .NEXT conference in Barcelona on May 21 2024, Nutanix announced power monitoring in the Nutanix Cloud Platform so operators can visualize and plan against actual power consumed by the hardware running their workloads. The data is based on measurements from the equipment in use and is updated in near real time. That means you can move from guesswork to decisions grounded in facts, right inside the same console where you already manage clusters.

What exactly did Nutanix add

The new capability lives in Prism, the management console for Nutanix Cloud Infrastructure. It brings node and cluster level power and energy usage into the same view as your favorite metrics like CPU memory and IOPS. You can see real time trends, review historical usage, and even pull the numbers via API if you want to feed cost and carbon models or your observability stack. In other words, one pane shows both performance and power so you can balance speed with efficiency in context. This announcement also builds on recent work in the Nutanix X Ray tool. X Ray now includes power and energy information alongside the usual performance stats. That lets you simulate scenarios and compare designs with power in the picture rather than treating it as an afterthought.

Why leaders should care right now

Sustainability is not a side project anymore. Nearly nine out of ten organizations say it is a priority and many are actively implementing initiatives as part of modernization. The pivot is toward being data driven about it. To be data driven you need data, which is exactly what this feature provides. For business leaders this means you can finally ask smart questions and expect defensible answers.

• What does this new AI service add to our monthly energy bill
• If we shift a noisy workload to a different cluster at night what efficiency gain do we get
• Can we prove that last quarter’s consolidation effort actually reduced power and not just rack count

With power in Prism you do not have to wait for a once a year facilities export. Your team can answer from the same system where they already move and scale workloads.

For architects and SREs this turns into practical superpowers

Let us make it concrete.

• Capacity planning with power included It is normal to size headroom for CPU memory and storage. Now add expected power envelope for cluster growth. If the envelope bursts, your team can show it with graphs your finance and facilities partners trust.
• Smarter workload placement When you see nodes that draw more watts per unit of work you can investigate. Are they older, sitting at poor utilization, or running chatty storage patterns You can right size or rebalance and then watch the curve flatten in the same dashboard.
• Data you can automate The API lets you pull power metrics into your favorite tools. Want a weekly cost and carbon summary by cluster Export the readings, multiply by your location’s grid carbon intensity and by your rate card, and publish an internal report.
• Better bake offs Use X Ray to compare designs and include power in the scorecard, not just latency and throughput. This is especially useful when you want to justify retiring a quirky legacy island that looks harmless on paper but eats watts for breakfast.

A quick refresher for the exec team

Here is a way I explain this during steering meetings without putting everyone to sleep.

• Power is the near term lever Embodied carbon matters, but the power you draw each day is what you can manage in real time. Visibility turns into action when you place or tune workloads based on what they actually consume.
• Carbon math is not mystical Take kilowatt hours from Prism. Multiply by your data center PUE to account for facility overhead. Multiply by the carbon intensity of your grid to estimate Scope two emissions. The last factor depends on location which is why workload placement across regions changes your footprint.
• You can simulate before you spend Nutanix provides an educational Carbon and Power Estimator that helps teams compare regions, PUE values, and platform choices so you can forecast before you buy or migrate. It is a helpful planning aid to complement real readings from Prism once you are live.

The feature in practice, step by step

Here is the rollout playbook I use with platform teams. One. Baseline first Turn on the power views in Prism and collect a clean baseline over two business cycles. Keep the context simple. Tag the clusters by business unit or application family so your reports are meaningful to leaders. Two. Pick three quick wins You do not need a moonshot. Go after noisy workloads that create obvious peaks. Typical candidates include batch data prep, test clusters left at full throttle, or VDI pools that never got their off hours policy. Move or tune, then watch the power line change. Three. Fold power into cost Create a very basic rate card. Dollars per kilowatt hour by site, plus cooling overhead via PUE, plus any demand charges if your utility uses them. Multiply by the readings you now have. When you present the next platform savings summary, include both infrastructure cost and energy cost avoided. Executives love a two line story that ends in clear savings. Four. Make it routine using automation Use the API to export readings on a schedule and drop them into your observability or cost management system. If you have a sustainability reporting tool, wire it up as well. Your future self will thank you when audit season arrives. Five. Use X Ray for design decisions When evaluating a new hardware generation or arranging clusters for AI and analytics, run X Ray tests and include power in the evaluation. The point is not to chase a single perfect number. The point is to compare designs with a repeatable method that includes the watts discussion by default.

What about results other customers see

Nutanix has long focused on consolidating resources with hyperconverged infrastructure to reduce footprint and energy use. On average, customers who shared experiences with Nutanix Cloud Infrastructure reported large reduction in physical footprint and significant reduction in energy consumption compared to their prior systems. Results vary based on age and configuration, but the direction of travel is compelling. The new power monitoring closes an old gap. It is one thing to tell a story about consolidation and expected efficiency. It is something else to show a graph from your own platform that proves the impact week by week and workload by workload.

Data plus culture makes this stick

A tool alone does not change practice. Here are cultural moves that make the difference. Make efficiency a feature request When a team asks for more nodes or storage, require a paragraph on expected power impact and the plan to optimize over time. With Prism power data in hand this takes minutes, not days. Publish a short weekly note Every Friday, send a three sentence note to stakeholders. What changed in the power curve. What you learned. What you will try next. The habit matters more than the format. Celebrate wins out loud When a small change saves real money or avoids emissions, post a screenshot and give credit. You will be surprised how quickly product teams start to offer ideas when they see others getting kudos for thoughtful placement and right sizing. Set a simple goal Pick something like ten percent reduction in non business hours power over the next quarter. It is specific, measurable, and invites everyone to look for small levers that add up.

A short guide for compliance and ESG teams

• You can attach Prism screenshots that show real readings to support narratives in your disclosures.
• You can feed raw numbers to your reporting system via API and demonstrate a consistent method.
• You can show leadership that modernization efforts are tied to measured results, not just hardware counts or survey claims.

What about AI and the wave of new demand

Everyone is asking how to make room for AI without blowing out power budgets. With these readings in hand you can do three useful things. Quantify the marginal watt Run an AI service in a controlled window and record the step change in power. You will know the real cost impact rather than guessing from a vendor spec sheet. Place with intent If you operate in more than one region, pair Prism readings with your utility rates and grid carbon intensities. A move from a carbon heavy region to a cleaner grid can pay back quickly. The estimator tool helps teams visualize those regional effects before you take the leap. Right size the rest AI often prompts a spring cleaning. While standing up new GPU friendly pools, you will spot under used clusters that idle high. Move or retire them, then show leadership that the AI spend was offset by eliminating waste elsewhere.

How this fits a broader sustainability toolkit

Think of three layers working together.

• Planning layer Use the Carbon and Power Estimator to explore scenarios by region, PUE, and platform choice. It is for education and planning before you build.
• Test and compare layer Use X Ray to compare architectures under load and include energy in the analysis. It is for proof during design and migration planning.
• Operate and improve layer Use Prism power monitoring to track real usage over time, automate exports, and drive continuous improvement with your teams. It is for the daily work of running the platform.

When you connect these layers, your plans line up with physics and your progress is visible to everyone who matters.

A friendly caution about expectations

Press releases rightly remind us that timing and exact features can change as products evolve. The Barcelona announcement framed power monitoring as the first step in a larger journey to give users more tools for sustainable IT. If you are reading this well after the date, check current release notes for details on what is generally available in your environment.

Getting started this week

You do not need a task force. Here is a lightweight plan that fits into a normal sprint. Day one Enable the power view in Prism and verify you see live readings by node and cluster. Confirm that your team can export the data through the documented API. Write down your current utility rate and PUE for each site. Day two Tag clusters by owner or application family. Pick one workload to tune. A simple policy such as lowering off hour reservations for a test environment often yields immediate savings without risk. Day three Build a basic report that shows power by cluster and by time of day. Share it with finance and your sustainability partner with a one paragraph explanation of what you plan to improve next. Day four Run an X Ray comparison for a single scenario you care about, like storage pattern changes or different node families, and include power in your conclusion. Present the results in a standing architecture review. Day five Publish your first weekly note with a small win. Keep it positive and specific. Momentum is the secret sauce.

Final thought from a fellow builder

For twenty plus years I have watched teams chase one percent performance gains while ignoring the power bill. That made sense when energy was cheap and compliance was simple. Today your platform is part of the sustainability story whether you lean in or not. Nutanix putting live power metrics into Prism makes leaning in easy. You will know what you draw. You will see what improves when you consolidate or tune. You will connect your daily decisions to company wide goals that include cost and carbon. And yes, you can have some fun along the way showing off before and after graphs that make your CFO and your sustainability partners smile. If you already run Nutanix, take the feature for a drive and measure something meaningful this week. If you are still exploring, review the announcement, skim the blog on Power Monitor, and try the estimator. Then ask your team the most important question of all. What would we do differently if we could see the watts

References

• Press release: Nutanix adds power monitoring to help customers track sustainability progress
• Tech blog: New X Ray power and metrics
• Blog: Power Monitor in Prism