Never miss a story from Corporate and Culture. Subscribe for in-depth analysis and articles.
Anyone designing a data center faces complex thermal management challenges. Yes, there’s a large amount of electrical power required, but the other side of that coin is that almost all the power gets turned into heat, putting a tremendous strain on the airflow and cooling systems. On top of the thermal issues, data center designers must deal with organizational issues, usually getting input requirements from several different places in the company. Consolidating these requirements and optimizing for the overall design is another huge challenge.
Obviously, discovering thermal problems after the data center is constructed is not good. Changes at that stage are costly. Couple that with the fact that once the data center becomes operational, there are changes over time to IT deployment that will vary from the intended white space design – this means that running an efficient data center with optimum performance becomes a delicate balance between capacity utilization, energy efficiency, and managing the risk of downtime.
The best approach is to build a digital twin of the data center that uses physics-based simulation, so a full analysis can be done entirely on the computer. This is similar in concept to the digital twins used in domains like automotive. It is very expensive to build a physical model of a car and put it in a wind tunnel. It’s much better to do all that modeling in a virtual environment with a digital twin.
But for a data center environment, there are certain attributes that need to be encompassed in the digital twin model to make it accurate and, therefore useful. Due to the complexity of airflow, cooling, and the need to keep each piece of IT at the desired temperature, data center digital twins must simulate airflow and cooling accurately using Computational Fluid Dynamics (CFD) calculations. For more information on creating a digital twin for a data center, please see our webpage.
There are several advantages to using a digital twin for data center design and operational management:
Whether designing a new data center, planning expansions for an existing site, troubleshooting problems, or designing critical power and cooling infrastructure, a digital twin provides confidence that the design will work without compromising the requirements. Using the power of physics-based simulation, a digital twin can dramatically simplify the design process to effectively test cooling and power performance.
Thésée DataCenter implemented a 6Sigma digital twin to use both in the design and operations of their data center. In partnership with HPE Services, they used a digital model to prove the design concept of their data center. Their data center digital twin was then used in operations to fine-tune performance and offer their clients a rich, innovative customer experience that became an essential tool in the communications between the data center and its customers. The digital twin allows client access to a 3D vision of their IT equipment, power, and operating conditions. Read case study.
Capacity planning is a huge challenge that starts with effective communication between teams and a planning process that is as agile as it is reliable. Often short-term fixes are prioritized over long-term capacity management strategies, reducing data center performance over the long run. A digital twin lets managers make informed capacity plans while still maintaining flexibility and speed. It encourages collaboration across teams, helps remove bottlenecks, and combats unnecessary costs.
A large healthcare enterprise uses a digital twin to modernize legacy data centers, minimize stranded capacity, and overcome power, cooling, and space challenges brought on by the extreme demands of high-density systems without compromising resilience. Read case study.
When building a data center’s digital twin, users can select from ready-configured intelligent library items from the IT servers to the ACUs, meaning that the small differences in behavior from one piece of equipment to the next will be represented in the overall data center digital twin model.
Wholesale and high-performance colocation provider Kao Data used a digital twin to validate its data center indirect evaporative cooling (IEC) design using CFD simulation software. Read case study.
A digital twin helps control resource utilization to realize business potential. A digital twin can help identify areas where savings can really add up. “What if?” analysis is much easier when it can be done on a computer. This not only helps with business costs but also contributes to running a more energy-efficient and sustainable data center.
Large global companies in automotive, healthcare, finance, and aerospace are using a data center digital twin solution to overcome power, cooling, and space challenges without compromising resilience. See our eBook here to see the impact: Read the ebook.
By using a digital twin, designers can ensure they meet service level agreement (SLA) requirements. A digital twin does not only express the current performance of a facility, but also its future potential. Different options can be compared before anything is built. Colocation providers can also use digital twin modeling to demonstrate to customers that proposed SLAs can be met.
Compass DatacentersTM uses a digital twin to ensure that SLAs remain unbreached. As a co-locator whose disaster-resistant, future-proofed data centers have met with critical acclaim, the company has used a digital twin from their first day to design and commission data centers that maximize revenue. Read case study.
A digital twin helps streamline efforts to manage the data center. With physics-based simulation, responses to new requests can be evaluated quickly and factually, saving time and money and reducing risk. This kind of modeling helps data centers to adjust to the evolving needs of the business. See what our panel of industry experts believe is provoking change in the data center industry and how the risk of change can be reduced. Read eBook.
For more information on using digital twins, here are some interesting articles: