#DataCenter Efficiency – Using #CFD Simulation to Optimize Cooling in Design & Operation


Modern Simulation Software for Data Centers. Source: Future Facilities


Energy is one of the biggest (if not the biggest) cost factors associated with data center operations, and represents the highest year over year growth rate. Unfortunately the efficient use of cooling can be like a game of Tetris. In Tetris, efficient use of space is impacted by the unpredictable shape of the blocks.  In the data center efficient use of cooling is impacted by the unpredictable airflow requirements of the IT equipment.  In Tetris, you can see the blocks and how they use the space.  But how can you do the same for cooling in the data center?

Are you confident the cooling optimization efforts have no negative impact on your data center operation and do not cause problems? Do you know if your current airflow is sufficient for the latest generation of server you plan to install? Can the cooling design for your facility still cope with today’s requirements of high-density deployments?

As today’s facilities have to be efficient and resilient it is a considerable advice to avoid trial and error strategies. State-of-the-art simulation techniques, such as Computational Fluid Dynamics (CFD), make the invisible visible, and validate the impact of IT infrastructure changes before putting them into action. CFD has become an essential tool for many companies as it allows users to quantify the airflow and temperature which would occur if physical alterations were made to the data center space.

Adapting new validation methods

CFD provides the capability to analyze every square inch of the data center, and determine the effectiveness of cooling within the racks and aisles. It also helps consider all the relevant aspects of cooling optimization with monitoring measures to validate simulation and planning results during operation.

The engineering simulation allows you to model any type of data center configuration whether it’s raised floor, slab, overhead cooling, in-row cooling, etc. Modern free-cooling technology can be incorporated such as direct and indirect evaporative cooling. You can even model complete control systems, hot-aisle or cold-aisle containment easily and compare each design variation. The simulation also allows you to analyze the impact of losing power to the entire facility (transient). Using CFD in the design phase is the best practice; today most sites are designed with the help of CFD tools in planning process by contractors. When the site is handed over to the user, CFD is usually no longer used on a regular basis – and that’s exactly where problems start to occur.

That’s why CFD is ideal to maintain to prevent changes for the worse. CFD can be employed when operators wish or need to check for a cooling perspective to make sure every piece of IT equipment is getting sufficient air flow at the right temperature in event of a failure. It has the capability to predict consequences of cooling failure.

Predict before you commit

CFD solves and even prevents many problems in data center design and operations. There are many benefits to CFD as there is no risk as changes are modeled and validated before action is taken. CFD integrates seamlessly into planning workflows and including it in operational procedures is nowadays a must for mature state-of-the-art data center management. Cooling optimization reduces energy costs, allows reclaiming of lost capacity, reduces downtime by preventing hotspots and optimizes space usage.

Modeling CFD allows effective communication between equipment suppliers, data center designers and operators. It is a risk-free way of experimenting within the data center to improve performance and capacity.

Though CFD modeling requires information about the size, content and layout of the data center to create a 3D model. If you are using a DCIM tool, the relevant data is already available at your fingertips and you just need to share it with the CFD tool.

An off-the-shelf adapter is available to connect FNT Command with Future Facilities 6SigmaDCX and share all changes between these tools. Integrating Engineering Simulation 6SigmaDCX with FNT Command IMAC Processes is a simple, 3-step planning process:

  • Run a simulation on your current planning scenario to visualize airflow and temperature
  • Look at the effects of the change that has been proposed
  • Cooling limits per cabinet can be committed back to FNT Command to facilitate further planning using internal threshold checks on the updated values.


Oliver Lindner, Head of Business Line DCIM at FNT, recently wrote an Expert Paper on this topic and explains in detail how to achieve performance improvements both in design and operation phase.

Download the expert paper titled here: Data Center Efficiency: Using CFD to Optimize Cooling in Design and Operation

This post originally appeared on the FNT blog: http://blog.fntsoftware.com/data-center-efficiency-using-cfd-optimize-cooling-design-operation/#more-489

Posted in Data Center DCIM Datacenter Datacenters Datacentre | Tagged , , , , , , , , , | Leave a comment

Airflow Management Can Help #DataCenter Operators Realize Robust Energy Savings

Upsite Technologies’ “4 R’s” Approach to Airflow Management Shown to Lower PUE and Increase Equipment Reliability

Holistic Methodology to Improve Computer Room Airflow Management Can Help Data Center Operators Realize Robust Energy Savings and Improve the Environment, as shown by new CFD Video by Future Facilities

Upsite Technologies, Inc., (Upsite) a leader in data center airflow management solutions, announced today that Computational Fluid Dynamics (CFD) modeling has demonstrated the energy savings outlined by its Video (below) 4 R’s of Airflow Management™ methodology. The company tapped Future Facilities North America (Future Facilities NA), the premier provider of engineering simulation software for data center design and operational planning, to demonstrate the findings through the utilization of its 6SigmaDCX CFD simulation tool.

Upsite’s 4 R’s of Airflow Management provides a guide for implementing changes to optimize cooling and achieve the greatest benefits of airflow management, including a lower Power Usage Effectiveness (PUE) score, reduced energy costs, and increased IT equipment reliability. The 4 R’s methodology details the improvements and best practices made to a data center’s racks, raised floor, rows, and room that will provide these benefits and optimize the cooling infrastructure. 6SigmaDCX was used to model a 4,000 sq. ft. data center and provide engineering simulation to assess the impact of the changes made to these four areas: rack, raised floor, row, and room. The simulation provided valuable information about how problem areas (e.g. hot spots) could be rectified and how capacity and operating cost benefits could be realized after making AFM improvements to each of the 4R’s. The execution of the steps resulted in:

  • Reduction in the maximum IT inlet temperature of 8.4° F
  • Cooling supply temperature increase of 10° F
  • Cooling unit fan speeds reduced by 35% and one cooling unit turned off
  • Partial PUE (pPUE) reduced from 1.54 to 1.34
  • Over $60,000 in annual savings for a 4,000 Sq. Ft. Data Center
  • 15 month ROI

“Given the many solutions available to improve data center airflow, the process of creating an effective airflow management plan can seem overwhelming,” said Lars Strong, Senior Engineer and Company Science Officer of Upsite Technologies. “Our 4 R’s of Airflow Management approach provides a clear strategy to optimize cooling and lower PUE. With the impressive results now demonstrated by CFD modeling, I anticipate that more owners and operators will be utilizing our 4 R’s methodology to accomplish this quickly, efficiently, and with a faster ROI.”


Original posted by BusinessWire

Posted in Data Center DCIM Datacenter Datacenters Datacentre | Tagged , , , , , , , , , | Leave a comment

Data Centers aren’t doing enough to be green

Most data centers are running well below capacity and hiding inefficiencies with green heat transfer initiatives

Posted by Ben Rossi on 4 May 2016 – InformationAge


The technology boom of the last ten years has seen people’s reliance on data centers increase exponentially. All of the online platforms that have made lives more convenient and connected, from streaming sites to instant messaging platforms, rely upon information stored in data facilities.

In the advent of trends such as the Internet of Things, this demand for data storage is set to skyrocket even further. However, we are already facing a critical problem: the environmental impact of data centers.

According to research from the Global Sustainability Initiative, data centers are now responsible for 2% of the world’s total greenhouse gas emissions. If the demand for data storage escalates at predicted rates, the volume of emissions that facilities produce will simply become unsustainable.

However, there is some good news – many of the problems with the current data center situation stem not from the sheer number of facilities in existence, but the ways in which they are operated.

Excess heat

Most data centers are currently running well below capacity. This inefficiency is often due to a fear of operational failure, and the belief that over-engineering is the only way to minimize this risk.

As a result, many data center operators are attempting to mask these inefficiencies through green heat transfer initiatives – using the excess heat to power public buildings such as swimming pools.

But this shortsighted strategy of ‘donating’ excess heat does nothing to address the underlying issues. Wouldn’t it be easier to solve the problem at the source?

Whilst the scale of new demand for data storage facilities is bound to require a certain amount of extra power, an ‘engineering gap’ currently exists, which prevents existing data centers from adapting to demands placed on them by businesses in the most efficient way.

As a result, the majority of data center energy expenditure via excess heat generation is avoidable. But there are techniques and tools that can be used on even the oldest data centers to reduce excess energy use and create truly ‘green’ facilities.

Green data centers

IT and facilities managers often over-provision and under-utilize the resources at their disposal, in order to avoid risk. In addition to being harmful to the environment, these inefficiencies also result in significant avoidable costs to the business.

Therefore, the challenge involved with making data centers greener lies in reducing operational inefficiencies without increasing the risk of downtime.

In part, this can be achieved through data center infrastructure management (DCIM) techniques. But if an operator is reliant on DCIM alone, the issue of the engineering gap remains.

The engineering gap in this instance refers to the disconnect between an IT manager’s prioritization of capacity, compared to a facility manager’s focus on efficiency.

The risk of making a change within the data center to improve efficiency is often, from an IT perspective, not worth the potential consequences should the system fail.

This means that the ability to predict the engineering impact of changes to the data center is potentially invaluable.

Predicting the impact of change on a data center requires a safe, offline environment to test proposals for change without the fear of disastrous consequences.

>See also: A change in tide: P.U.E vs. W.U.E and the future of the green data centre

One of the best ways to do this is to use engineering simulation. Through the use of 3D modelling to represent the data center, power system simulation (PSS) and computational fluid dynamics (CFD), it is possible to identify whether a data center is operating with wasted capacity, and therefore excess heat production as a result.

Implementation of these methodologies means businesses can drive up utilization, thereby minimizing excess heat production.

An added bonus is that engineering simulation can be used to assist with installing high power density IT into a legacy facility, resulting in a technology boost for business processes, without the need for additional data center construction and unnecessary heat generation.

Rather than worrying about ways to recycle excess heat produced by facilities, businesses can instead conduct an evaluation of the ways in which they can be optimized.

Sourced from Jon Leppard, director, Future Facilities

Posted in Data Center DCIM Datacenter Datacenters Datacentre | Tagged , , , , , , , , , | 1 Comment

Uncovering the Value of Calibrating #DataCenter #CFD Models

Why should you care about engineering simulation and calibration in your data center?


Future Facilities has teamed up with the Center for Energy-Smart Electronics Systems (ES2) to uncover the true value of calibrating a virtual facility model from thermal and airflow perspectives.

Video Gallery        Calibration Paper        ES2 Website        Technical Papers

Posted in Data Center DCIM Datacenter Datacenters Datacentre | Tagged , , , , , , , , , , | 1 Comment

Your #datacenter can’t change fast enough

Originally posted in DatacenterDynamicsoptimized

How do you create data centers that can respond to business needs with speed, control and visibility?

Jon Leppard Jon Leppard

The speed and resilience of a data center is something we obsess over – making efficiencies and improving performance is what makes working in IT so satisfying. We live for the next tech breakthrough that can process more, store more, or just work harder on less power. Luckily, with Moore’s law (well, roughly), we’ve enjoyed a steady supply of these breakthroughs over the past decade.

But, when the pressure is on to make a change, most data centers are slow. There’s a lot to consider, especially when we are talking about larger facilities with mission critical workloads, multiple systems, failsafes, cooling setups etc. to take into account.

Do we want fast change?

Pit stop

The speed of business is only increasing – the successful launch of in-memory analytics such as SAP HANA proves that businesses are taking competitive advantage not just from big data, but fast data. Time is, more than ever before, money. The trend over the near term appears to be hardware struggling to keep up with the demands placed upon it by new software workloads.

When the pressure therefore inevitably comes down to begin migration to faster, more efficient data centers, the DC team’s ability to control the risk of the process is far too limited. When varied workloads are pushed into the data center by the business, or something needs to change within the facility itself, it poses questions about capacity and resilience which need very precise answers in order to avoid lengthy and costly mistakes.

Can we change fast?

The reality for most data center managers is that answering questions like do we have the capacity? or are we exposing ourselves to too much risk? is often a combination of historic trends, experience and a bit of educated guesswork. Therefore, quite rightly, the IT team will appeal for time from the business to assess and get things as close to “right” as possible – who can blame them? The data center is dealing with mission critical applications, SLA’s and customer data, failure is unacceptable.

Right now, the facilities and IT teams in this situation simply can’t change fast. But businesses don’t have the time to wait on the data center. The collective data center industry is doing what it’s done for the past decade – it’s waiting for a technological breakthrough to save the day.

Is there a solution on the horizon?

Yes, there certainly is! New designs for software-defined, homogenized facilities look like they will be going a long way towards responding to this need. The only problem here is that for most organizations (those without ludicrously large IT budgets) the software-defined data centers are between five and 10 years away.

That’s a bit beyond the scope of the timeline most businesses are setting their data center teams to implement change, sadly.

We therefore need to create data centers that can respond to these business needs with absolute control and visibility, to remove risk for the equation. We call this concept the ‘Fluid Data Center’.

What’s a Fluid Data Center?

Rather than an amazing new cooling system, the Fluid Data Center is a concept where capacity and risk can be accurately and quickly snap-shotted. A Fluid Data Center can “pour” my resource towards either end of this spectrum with safe knowledge of the impact this will have on either the capacity or the resilience of the entire facility.

It can do this on a case by case basis, and can do this quickly.

It’s achieved by knowing exactly what is happening currently within a data center and then using advanced engineering simulation tools to map out what the impact of any given change would be. Not just in terms of power draw, but on what the impact would be on the air flow of a room, the additional strain on a given AC unit etc. down to the fine detail.

What this tends to result in, aside from happier business teams, is incredibly efficient data centers. At the moment the only solution to not knowing precisely the limit of a DC is to factor in a healthy safety margin – this could be an extra AC unit or two, in simple cases. A fluid data center has these turned off – bringing down PUE – and is able to smartly communicate with the business that these will be turned back on if X occurs.

A Fluid Data Center knows exactly how much juice it has, and the size of the container – and it uses this information to act faster and safer than human predictions could ever achieve.

But the best bit about it is that it’s a solution to a growing problem that is available right now, instead of on the horizon.

Jon Leppard is a director at Future Facilities, a company that specializes in engineering simulation tools.

Posted in Data Center DCIM Datacenter Datacenters Datacentre | Tagged , , , , , | Leave a comment

Future Data Centers are Dynamic…but what about today’s Facilities?

Screenshot 2015-12-04 09.30.45

Future models for the data center are highly dynamic; workloads easily transferred and managed across a highly homogenized facility, that is effortlessly managed by sophisticated software. Or at least so the prevailing concepts within the industry would suggest. But what does this mean for today’s data centers – facilities that may be in operation for ten years or more? This question is answered by Matt Warner, Development Manager at Future Facilities.

The Future of Data Centers, Delivered Today

I have a goal for every data center manager: make your facility more responsive, risk-free and predictable. It sounds like a lot to ask, but it’s an achievable goal. I know because I’ve seen a lot of businesses do it.

This vision for better data centers in the here and now is encompassed within what we at Future Facilities call ‘The Fluid Data Center.’ However our industry is not inherently fluid, and there are a great number of data centers out there yet to make this change. Instead, most data centers exist in a different state.

When we speak to IT and facilities managers across our industry, there is a recurring pattern – current facilities are being managed as what we call ‘Static Data Centers.’ Before I explain what this means, it’s important to note that this is not a criticism. The reality is that there are a huge number of pressures in the management of the data center that traditionally have all but demanded that facilities are ‘static.’

So what is a static data center? Typically it’s characterized as a facility in which there is a reluctance to make changes, and where new deployments or reconfigurations of hardware or software are slow and risky. A static facility is usually managed either looking backwards to past trends, or forwards with the educated guesswork of experienced data center professionals. It is all but impossible to make this sort of facility highly flexible while remaining absolutely resilient.

Leaving Static Facilities Behind

These static data centers are in so many ways a legacy. Historically, the incremental purchase and installation of hardware, initiated in response to erratic demands from the business, was the norm. The last ten years or so have seen this situation become far more complex for most businesses. Their data centers are diverse, with a wide variety of installed technologies, frequently leading to fragmentation.

This conspiracy of factors has results in two overriding, and highly detrimental trends:

  1. Most IT teams have over-provisioned and under-utilized within their data center to safeguard the delivery of compute power for which they’re ultimately responsible. This strategy has been deployed to ensure that they don’t allow their applications to fall over
  2. Facilities teams, similarly, have made their own sacrifices to safeguard the resilience of their mission critical facility. Typically they have over-engineered and over-cooled to minimize downtime

So how do we make a data center more ‘Fluid’? We start by reminding ourselves why – outside of the operational parameters and SLA to which we’ve agreed – we care about making our data center more flexible. Ultimately it comes down to business facilitation.

A Critical Change

The data center is often referred to as a ‘critical facility’ within the organization. Indeed today its role is more important than ever. The business itself is subject to more rapid changes as a result of rapidly advancing technology, demands from consumers for immediate responses and hugely competitive industry landscapes. This is an era of unparalleled interaction between lines of business and technology, and the demands on the data center are only increasing. Hence the idea of a ‘safe but static’ data center (which in honesty was never a reality anyway) becomes anachronistic. There’s just no place for it today.

What’s more, in the face of this more dynamic demand, budget pressures on IT and Facilities remain high. A static data center configuration is wasteful on resources. Therefore the cost of compute ($/W) results in a correspondingly high cost of delivering business outcomes. These sorts of difficulties may not arise in the cutting edge of Marketing discussions or at the latest Sales conference, but rest assured that the CIO and the Board upon which he sits are looking at those numbers with watchful eyes.

Becoming Fluid

Gone is any remnant of “keeping the lights on at any cost.” Today the mission is “optimizing how to keep the lights on.” And optimizing means combining both performance and risk management.

On an operational level, an efficient data center is one where power and cooling supplied by the Facility balances the IT demand. In more commercial terms, the data center must also have the ability to maintain this balance while being completely flexible to the needs of the business. That means changing things – often, and with no risk of downtime.

In the current model of the data center, this doesn’t work, primarily because of a decision making gap between IT and Facilities. The issue lies in the fact that both are currently operating independently. Many organizations have deployed DCIM technology with a goal of crossing the data and process gaps that are found within any data center facility. This is a positive step, but doesn’t cover all bases. In fact, the majority of facilities today are managed whereby the operator makes decisions without any clear insight into the engineering impact they may have on the other side of the gap. In other words, IT doesn’t know how it will affect Facilities, and vice versa.

Data centers today therefore suffer from an ‘engineering gap.’

We Must Close the Engineering Gap

This engineering gap isn’t just a nice buzzword. It’s a real thing. And a real problem too. Within almost any facility you choose to inspect, you’ll find this gap, and it’s exposing these organizations to risk of:

  • A loss in business performance – expressed as a loss of hardware Availability
  • Wasted CapEx – the direct result of stranded Capacity
  • An unnecessary increase in OpEx – occurring due to loss of Cooling Efficiency
  • The end point for the business is that these issues result in Increased Cost

You can take solace however from the fact that many organizations have overcome these issues. There does NOT have to be an engineering gap. While we cannot map all of the complex process at play in the data center on a scrap of paper, in our heads or even in a traditional DCIM tool, we can predict them using engineering simulation. Through engineering simulation it’s possible to create a 3D model of the data center, simulate the power systems and run computational fluid dynamics modelling to predict cooling.

Example: Deploying High Density IT & Engineering impact



Change – Made Safe

This process of engineering simulation gifts to the data center manager an apparently magical tool. It gives them a safe, off-line environment in which they can explore and test the changes required within the data center. So when the demands of the business come flooding in, resulting in huge variation in the workloads being handled, the data center can perform with absolute resilience. Change is no longer the enemy.

And so we have introduced one of the great challenges of our industry today, along with a goal that any of us can achieve. It’s also important to remember that this isn’t just a vision of the future; it’s a vision for today, and we call it The Fluid Data Center.

First posted on Data Centre Network 

Posted in Data Center DCIM Datacenter Datacenters Datacentre | Tagged , , , , , , , , , | Leave a comment

Applying the Scientific Method in Data Center Management


Data center management isn’t easy. Computing deployments change daily, airflows are complicated, misplaced incentives cause behaviors that are at odds with growing company profits, and most enterprise data centers lag far behind their cloud-based peers in utilization and total cost of ownership.

One reason why big inefficiencies persist in enterprise data centers is inattention to what I call the three pillars of modern data center management: tracking (measurement and inventory control), developing good procedures, and understanding physical principles and engineering constraints.

Another is that senior management is often unaware of the scope of these problems. For example, a recent study I conducted in collaboration with Anthesis and TSO Logic showed that 30 percent of servers included in our data set were comatose: using electricity but delivering no useful information services. The result is tens of billions of dollars of wasted capital in enterprise data centers around the world, a result that should alarm any C-level executive. But little progress has been made on comatose servers since the problem first surfaced years ago as the target of the Uptime Institute’s server roundup.

Read more: $30B Worth of Idle Servers Sit in Data Centers

One antidote to these problems is to bring the scientific method to data center management. That means creating hypotheses, experimenting to test them, and changing operational strategies accordingly, in an endless cycle of continuous improvement. Doing so isn’t always easy in the data center, because deploying equipment is expensive, and experimentation can be risky.

Is there a way to experiment at low risk and modest cost in data centers? Why yes, there is. As I’ve discussed elsewhere, calibrated models of the data center can be used to test the effects of different software deployments on airflow, temperatures, reliability, electricity use, and data center capacity. In fact, using such models is the only accurate way to assess the effects of potential changes in data center configuration on the things operators care about, because the systems are so complex.

Recently, scientists at the State University of New York at Binghamton created a calibrated model of a 41-rack data center to test how accurately one type of software (6SigmaDC) could predict temperatures in that facility and to create a test bed for future experiments. The scientists can configure the data center easily, without fear of disrupting mission critical operations, because the setup is solely for testing. They can also run different workloads to see how those might affect energy use or reliability in the facility.

Read more: Three Ways to Get a Better Data Center Model

Most enterprise data centers don’t have such flexibility, but they can cordon off sections of their facility as a test bed, as long as they have sufficient scale. For most enterprises, such direct experimentation is impractical. What almost all of them can do is create a calibrated model of their facility and run the experiments in software.

What the Binghamton work shows is that experimenting in code is cheaper, easier, and less risky than deploying physical hardware, and just about as accurate (as long as the model is properly calibrated). In their initial test setup, they reliably predicted temperatures with just a couple of outliers for each rack, and those results could no doubt be improved with further calibration. They were able to identify the physical reasons for the differences between modeling results and measurements, and once identified, the path to a better and more accurate model is clear.

We need more testing labs of this kind, applied to all modeling software used in data center management, to assess accuracy and improve best practices, but the high-level lesson is clear: enterprise data centers should use software to improve their operational performance, and the Binghamton work shows the way forward. IT is transforming the rest of the economy, why not use it to transform IT itself?

About the author: Jonathan Koomey is a Research Fellow at the Steyer-Taylor Center for Energy Policy and Finance at Stanford University and is one of the leading international experts on the energy use and economics of data centers.

Sign up for Jonathan Koomey’s online course,Modernizing Enterprise Data Centers for Fun and Profit. More details below.

Originally published on Data Center Knowledge 


Posted in Data Center DCIM Datacenter Datacenters Datacentre | Tagged , , , , , , , , | Leave a comment