Key Challenges in Data Center Design & How to Solve Them

A survey by the Uptime Institute of more than 800 data center facilities worldwide has shown that data centers are aging and becoming obsolete. Unfortunately, companies are neither planning ahead nor demonstrating prompt actions in expanding or modernizing the existing data centers. Owners of 38% of the data centers surveyed said that their current data centers were built more than 4 years ago, reflecting the challenges so many organizations have while coping with the existing data centers’ intense power and cooling demands.

Between  2020-25, global data creation is forecasted to grow by 180 Zettabytes! So apart from data centers becoming old and obsolete, the growing storage needs, resulting in overcrowding of data centers, is a major problem for the existing data centers. All of this makes it imperative to expand and modernize the current data centers.

Unfortunately, while modernizing or expanding their data centers, most firms fail to assess their needs properly, leading to faulty and subpar designs that fail to meet the required performance goals or business needs.

5 Key Data Center Design Challenges and How To Mitigate Them

Data center owners face several challenges during the design process. Here we will discuss the key design challenges companies face during their data center expansion and how to mitigate them. This will help you avoid the common pitfalls and with flexible, modular options, making it possible to future-proof your data center along with meeting your current business goals and budget.

1. Failure To Take The Total Cost Of Ownership (TCO) Into Account During The Initial Design Phase

Most companies fall into the trap of focussing solely on capital cost. Capital cost modeling is an indispensable part of any design process. However, if you do not take into account the cost to operate and maintain your business infrastructure, it will severely affect your current business processes and planning.

The Solution:

There are primarily two ways to build a data cost modeling- the maintenance cost and operating cost.

Maintenance Cost: Maintenance costs are the costs associated with the proper maintenance of all the supporting infrastructures. They include OEM equipment maintenance, data center cleaning costs, and subcontractor costs for various repairs and upgrades.

Operating Cost: Operating costs are associated with the daily operational and on-site personnel costs. They include staffing levels, safety training, skill training, capacity management, and other QA/QC procedures.

If there is a failure in calculating a long-term operational and maintenance budget, building an optimized ROI Model becomes increasingly challenging to support intelligent business planning.

If your objective is to build and expand a business-critical data center, the best approach is to focus on these parameters: 

  1. Capital Expense

  2. Operational and Maintenance Expense

  3. Energy Cost

If any of these crucial components are left out, the model will not perfectly align with the business goals and risk the entire management process.

2. Poor Cost-To-Build Estimation

Another common mistake in the design process lies in the estimation process itself. Financial requests made to the board of directors for capital needed for the growth and expansion of the data center are set too low, which often results in poorly planned data centers.

The Solution:

An ideal flow in the decision-making looks something like this:

  • A holistic TCO approach is adopted, and a cost-effective financial model is devised. This helps arrive at an accurate approximation of the actual cost.

  • The budget is sent to the board of directors for approval. It is then investigated, and a proper final budget is decided.

  • Invest enough time in planning how to spend wisely from the approved budget.

3. Improperly Setting Design and Performance Criteria

The criteria that often plummet the design process down the drain are – First, everyone wants a Tier 3 design even though not everyone needs it. Second, most metric systems of kilowatt per square foot are not supported to meet a company’s business goals.

The Solution:

The key to avoiding this problem is– Try not to overbuild. It’s just a waste of the working capital of your business. Higher tier facilities also lead to higher operational & management costs and require a more significant amount of energy.

The perfect roadmap would be to:

  • Establish the most optimized business model and optimize your performance by thoroughly analyzing your business goals, risk protocols, and business requirements.

  • Set your capital expenses and operational costs around them.

  • Finally, make sure you fix your financial model right before approaching the board of directors with the final budget.

Following these steps will help you curb this common mistake and save your business a significant amount of working capital.

4. Selecting A Site Before Design Criteria Are Devised

Organizations frequently make the mistake of searching for the perfect space to build a data center even before having their design and performance criteria in place. The major problem arising from selecting a site even before the design criteria are in place is that it does not often meet the design requirements.

The Solution:

So while designing data centers, it is necessary to consider several factors such as power availability and cost, natural issues such as earthquakes, tornados, floodings, etc. In addition, baseline users must have business models that have the inherent need to build or renovate a shell in their core business region.

5. Overcomplicated Designs

The data center is a complicated system with multiple moving parts; Power, cooling racks, cabling, IT, and many more. Usually, they are manufactured separately and assembled together. This results in forming an overcomplicated design due to the large number of parts involved. We flip a switch and hope it works. However, it often doesn’t.

Designing an effective system regardless of the target tier chosen is the key to create an optimized design with lower costs. A simple design has fewer parts that do not compromise the performance and can be operated easily.

The Solution:

The data center design must be kept simple. The benefits of a simple design are as follows: 

  • Complex designs have more parts and components, thus more failure points. This also increases the probability of equipment malfunctioning.

  • Complex systems increase risks. These can be operational risks like failure to operate parts and components safely or maintenance issues arising from human error.

  • A simple design often demands lower operational and maintenance costs. Lesser pieces of equipment reduce the maintenance costs compared to complex structures with more parts.

Data Center Design Suggestions From Mechartes

Although many data centers face these challenges in their designing process, you can easily avoid them. Here are the tips curated just for you by the experts here at Mechartes.

Efficient Design

You should choose power systems (i.e., UPSs, PDUs) specially designed for high efficiency. You must also choose cooling systems that allow for maximum economic cooling hours within your geography. You must also pay close attention to the design choices that impact air distribution (containment systems, raised floors, etc.), as those directly impact the efficiency of the data center.

Connected Systems

Select systems that use cloud-based and mobile-friendly data center infrastructure management (DCIM) software that can be managed remotely. This will allow you to leverage information for predictive maintenance instead of traditional calendar-based maintenance. This will help maximize the life of components while managing their availability due to optimized management of resources.

Redundant Systems

If you need high activity to process a large amount of data, you should design keeping redundancy in mind. This is a backup for the system, so that component failure or maintenance activities don’t overload the critical load.

Scalable Designs

Future loads are often unpredictable. So having an easily scalable system is an important attribute for a data center. Many UPSs are modular, so as the load grows, additional power modules can be easily added. You must also consider the current and future density needs. As chip density grows, can your cooling system keep up? There is an increasing trend towards liquid cooling as chip densities climb above the threshold limits of air-cooled data centers and are easily scalable.

At Mechartes, we focus on providing accurate simulation results with a professional approach using advanced engineering tools. We provide  Data Center Validation Services, including the pre-design stage, design stage, and construction stage.

Our specialty includes Data Center Architecture and Engineering. The different stages we design and the specific technologies we use are:

  • Pre-Design Stage: We use External Flow CFD analysis for the chiller and generator yards present in the Data Center sites. We carefully assess the wind directions and weather conditions to judge the perfect placement of the chiller and generator units.

  • Design Stage: We use CFD Analysis to validate and optimize the HVAC (Heating, Ventilation, and Air Conditioning) design of the following areas as per specific project requirements: Data Hall, Generator room, and DRUPS room.

  • Construction Stage: Stress analysis is used to analyze the existing piping networks as it helps in analyzing the compatibility of the design with its weight, pressure, and thermal stress. Based on the detailed stress analysis report, we provide support and design suggestions for the most effective design.

For expert consultation on designing or expanding your data center, visit Mechartes and get the best guidance adhering to all the common problems beginners face, all the way up to advanced tier 3 designs of data centers for mega users.

Published On : October 21, 2024

Leave a Reply

Your email address will not be published. Required fields are marked *