Most data centre cooling studies look inside the building: the data hall, the racks and the cooling units. But every cooling system rejects its heat outdoors and draws in outdoor air to do it. A data centre microclimate simulation looks at that outside air, because if it is hotter than the design assumed, the cooling plant loses capacity and efficiency before a single server is affected.
In Singapore, where many data centres sit on tight sites next to other buildings, this can go unnoticed because most studies stop at the building wall. An external microclimate study shows what the air around a facility is actually doing.
What is a data centre microclimate simulation?
A data centre microclimate simulation is a Computational Fluid Dynamics (CFD) study of the air immediately around a data centre building. It models wind, heat and exhaust across the site to show the air temperature arriving at each chiller, condenser, cooling tower, generator and fresh air unit, which can differ noticeably from the weather station reading.
A microclimate is simply the local air condition around one building. For a data centre, the part that matters is the air at the equipment intakes, because that is the air the cooling plant has to work with.
Owner's reply
Thank you, Jack, for the kind words. We are glad our team could support you quickly, and we truly appreciate you taking the time to share your experience. Looking forward to working with you again.
Owner's reply
Thank you, Kenny, for your kind review and recommendation. We truly appreciate your trust in Megagenix and are glad that you found our support responsive, reliable and straightforward. It has been a pleasure working with you, and we look forward to supporting you again on future projects.
Owner's reply
Thank you very much for your kind review and recommendation. We are pleased to know that you found our CFD analysis technically strong and our reports clear and easy to understand. It was a pleasure supporting your project, and we look forward to working with you again in the future. ❤️
Owner's reply
Thank you Mr Choong for your kind words and recommendation. We truly appreciate your trust in our team. We understand the project timelines can often be challenging, and we always do our best to support our clients by responding promptly while maintaining the quality of our work. At Megagenix, we believe our role is not only to provide simulation results, but also to help clients make informed engineering decisions. We look forward to working with you again on future projects.❤️
Owner's reply
Thank you, Wen Hong, for your 5-star review! We truly appreciate your support and are glad you had a positive experience with Megagenix.
Owner's reply
Thank you very much for your kind feedback and for giving Megagenix the opportunity to support your project. We are pleased to know that our responsiveness, clear communication and CFD analysis helped your team evaluate the design changes and make decisions more efficiently. We truly appreciate your trust and look forward to supporting you again in future projects.
Owner's reply
Hi Jason, Thank you very much for your kind and thoughtful review. We truly appreciate your continued trust in Megagenix and the opportunity to support you across several data centre CFD projects. It has been a pleasure working with you and your team to overcome project challenges and meet the project timelines. Your feedback is greatly appreciated and motivates our team to continue delivering responsive and reliable technical support. We look forward to continuing our collaboration and supporting your future projects. ❤️
Owner's reply
Thank you, ColdTec team, for your kind review and trust in Megagenix. We are pleased that our team was able to support your urgent Heat Load Calculation and CFD analysis within the required timeline. We truly appreciate your recognition of our responsiveness, professionalism, and technical expertise. We look forward to supporting ColdTec again on future engineering projects.❤️
Owner's reply
Thank you, Mr. Rentatt for taking the time to share your experience and for your kind recommendation. We truly appreciate the opportunity to support DCD. Your trust means a lot to us, and we are glad that our team could provide responsive and practical engineering support throughout the project. We look forward to collaborating with you again on future CFD and engineering simulation projects. ❤️
Owner's reply
Thank you, Amy, for your wonderful feedback. We are pleased that our team was able to provide practical recommendatkons that helped your team to address the hot spots early and potentially reduce downstream modification costs. At Megagenix, we strive to bridge the gap between simulation results and real-world engineering solutions. We really appreciate your trust and look forward to supporting your future projects.😊
What affects the air around a data centre?
The air around a data centre is shaped by the facility itself and by everything near it. Hot exhaust from its own equipment, cooling tower plumes, neighbouring buildings, plant screens and changing wind directions can all raise the temperature at the intakes, often in ways that internal data hall studies cannot show.
| Factor | What happens | Equipment most affected |
|---|---|---|
| Hot air recirculation | Hot exhaust from condensers, dry coolers or generators drifts back into the intakes of the same or nearby equipment. | Air-cooled chillers, condensers, dry coolers |
| Cooling tower plumes | Warm, humid air from cooling towers is drawn back into the towers or into neighbouring air intakes. | Cooling towers, fresh air units |
| Neighbouring buildings | Adjacent buildings block wind, trap heat or push their own exhaust towards the plant. | All rooftop and ground-level plant |
| Roof and plant screens | Screens installed to hide equipment can also stop hot air escaping. | Rooftop condensers and chillers |
| Wind direction | Singapore’s prevailing winds shift between the northeast and southwest monsoon seasons, so a layout that works for part of the year may struggle for the rest. | Any plant that relies on wind to clear its exhaust |
Why does site airflow affect cooling performance?
Cooling equipment is rated for a specific intake air temperature. When the air reaching it is a few degrees hotter than designed, the equipment has to work harder to reject the same heat. Air-cooled chillers and condensers lose capacity and draw more power, and cooling towers lose effectiveness when they take in their own warm plume.
On the hottest afternoons, when the IT load and the outdoor temperature are both high, that lost capacity can push the plant close to its limit. It also shows up in the facility’s Power Usage Effectiveness (PUE), the ratio of total facility energy to IT energy, because the cooling plant spends more energy than it should. Our PUE optimisation page explains how a facility can be modelled against its PUE target.
What does an external microclimate study show?
An external microclimate study shows the intake air temperature at every key piece of plant, where hot exhaust and plumes travel, which wind directions are worst for the site and how proposed changes would perform. Because it is a model, layouts and fixes can be compared before anything is built or moved on a live roof.
| Output | What it tells you |
|---|---|
| Intake air temperatures | How the air at each chiller, condenser, cooling tower and generator compares with the design assumption |
| Exhaust and plume paths | Where hot air and plumes travel, and which equipment they reach |
| Worst-case wind directions | Which conditions stress the site most, not just the average day |
| Effect of changes | How options such as raising a stack, moving a unit, changing screen design or adding a baffle would perform |

Similar methods apply beyond data centres. The published generator set CFD case study modelled how heat from a genset spreads to a nearby building, using a maximum ambient temperature of 38.2°C and a wind direction based on the direction of maximum wind velocity at 10 metres, and checked whether hot exhaust short-circuited back into the inlets.
When should you commission a microclimate study?
A microclimate study is most useful at design stage, when plant layout and screens can still change at low cost. It is also worth doing before adding plant to an existing site, when a neighbouring building is built or altered, or when the cooling plant underperforms on hot days for reasons that internal data does not explain.
| When | Why it helps |
|---|---|
| Design stage | Plant layout, spacing and screen design can still change at low cost. |
| Before adding capacity | New chillers, condensers or generators placed near existing ones can affect each other’s intake air. |
| When nearby buildings change | A new or altered building can change the wind around the site. |
| When the plant struggles on hot days | The cause may be outside the building, where internal data hall studies do not look. |
An Example Scenario: The Plant That Struggles Every Afternoon
An operator adds new air-cooled condensers to the roof of an existing facility to support an IT load expansion. The units go into the only free space, next to the existing plant and behind a tall screen that hides the equipment from the street. On paper, the new capacity is enough.
In operation, the plant runs close to its limit every afternoon. Intake temperatures at some condensers are several degrees above the outdoor reading, and the problem is worse in one monsoon season than the other. The internal data hall study shows nothing wrong, because the problem is outside.
A study for a site like this would typically follow these steps:
| Step | What is done |
|---|---|
| 1. Measure | Record intake temperatures at the affected units and collect the site’s wind and weather data. |
| 2. Model | Build an external CFD model of the roof, the plant, the screen and the surrounding buildings. |
| 3. Run wind cases | Run the model across the site’s key wind directions to find where hot exhaust is recirculating and why. |
| 4. Test fixes | Compare intake temperatures for practical options such as screen openings, discharge extensions or relocating units. |
| 5. Recommend | Choose the option that restores capacity at the lowest cost and disruption. |
Problems like this are usually far cheaper to test on a model than by trial and error on a live roof.
Talk to an engineer about your site
Megagenix is an independent engineering consultancy offering data centre CFD in Singapore and Malaysia. If your cooling plant underperforms on hot days, or you are planning new plant on an existing site, speak to our engineers about a data centre microclimate simulation. You can also read more about our data centre CFD services and PUE optimisation services.
Frequently Asked Questions
Is a microclimate study the same as a data hall CFD study?
No. A data hall CFD study models the air inside the building, around racks and cooling units. An external microclimate study models the air outside, around the plant and neighbouring buildings. The two answer different questions, and together they show whether the whole cooling chain is working as designed.
What is hot air recirculation?
Hot air recirculation happens when hot exhaust from cooling or power equipment is drawn back into the intakes of the same or nearby equipment. The intake air is then hotter than designed, which reduces the equipment’s capacity and efficiency and can push the plant towards its limit on hot afternoons.
Does external airflow affect data centre PUE?
It can. When intake air is hotter than designed, the cooling plant uses more energy to reject the same heat, which raises the facility’s PUE. How much depends on the plant type, the layout and the site’s wind conditions, which is why it is worth modelling rather than assuming.
When should a microclimate study be done?
A data centre microclimate simulation is best done at design stage, when layout changes are cheap. It is also valuable before adding plant to an existing site, after nearby buildings change, or when the plant underperforms on hot days without an obvious internal cause. Earlier studies usually leave more low-cost options open.



