Applied in almost all industries that use design software, engineering simulation enables a thorough testing process through the numerical simulation of a product or building’s physical properties in different conditions and environments. Simulation, known in the industry as computer-aided engineering (CAE), can be used early in the development process — from the first design version — reducing the number of physical prototypes required.
Overview
Simulation is not intended to replace physical prototypes altogether but rather to serve as a complement, enabling an iterative design process, saving time and costs.
The standard workflow of a simulation, which is usually a fluid dynamics simulation or a structural analysis, includes meshing the design (created in a CAD software), adding boundary conditions, setting up the analysis parameters, running it, and post-processing the results. The latter step provides the information needed in order to make design changes and improve performance or durability.
Despite its broad range of potential use cases, engineering simulation is not yet part of every engineer’s toolkit. This is due to the barriers traditionally associated with on-premises CAE software, including the high upfront investment, required software, and specialized training.
Five Benefits of Cloud-Based CAE
Cloud-based CAE has helped remove the barriers associated with the technology, fostering its adoption by a broader range of users across industries. Following are the five main benefits it brings to development processes in general, including data center design.
1) Lower costs
Cloud-based CAE solutions are generally delivered as software-as-a-service (SaaS) models, which imply a subscription fee. This is more affordable than buying a perpetual license and allows users to pay only when the technology is needed. Moreover, they can be accessed from any laptop or Wi-Fi-enabled device. For traditional on-premises software, numerical simulations are computationally intensive, so special hardware is required in order to run them.
2) Easy to learn
Cloud-based CAE solutions can be used by professionals who don’t have a background in engineering simulation or high-performance computing (HPC) systems. This technology does require learning, but the training period is short.
3) No installation, no maintenance, no overhead
Since it’s a web-based application, software installation, maintenance, and upgrades are no longer required. Product updates still happen regularly, but they are seamlessly integrated into the solution, so users have access to the latest version at all times.
4) Flexibility
Cloud computing allows for scaling resources up — by parallelizing arbitrary amounts of simulation runs — and down — when the workload is lower.
5) Collaboration features
The need for teamwork or asking for feedback from managers or clients has always been present. With the recent shift toward home office and remote work, however, it has become top of mind for most companies and professionals.
![cloud-based CAE software](/ext/resources/images/2020/CFD-Collaboration.jpg)
In engineering, cloud-based solutions — not only for CAE — enable team collaboration with access rights and easy sharing options. This grants all stakeholders of a project to access that project, the simulation runs, and related results via a click of a button.
Data Center-Specific
The benefits of CAE in general — and cloud-based solutions in particular — apply to most industries and applications. But why is it that important for data center designers?
Actually, data centers are one of the verticals that require engineering simulation the most due to the high energy efficiency and cooling requirements of the industry. A typical data center can consume as much energy as 25,000 households.
The energy consumption for storing and maintaining large amounts of data can be reduced with optimized infrastructure. For this, temperature management is key. With the equipment running constantly, cooling systems are essential in data centers. But these systems also consume energy, sometimes even more than the servers they need to cool. A well-designed cooling system is therefore a must in data centers.
In thermal management, engineering simulation, and computational fluid dynamics (CFD) in particular, is a standard technology. CFD allows engineers to test data center designs and their cooling systems in order to identify potential issues or optimization opportunities. Simulation results enable the visualization of airflow patterns and predict the flow and temperature distribution, giving the required information for optimizing the supply air temperature and the supply airflow rate, reducing the overall cooling costs.
For example, the two configurations in Figure 1 have the same inlet and outlet conditions, but the inlet position differs. This might seem like a minor change, but in reality, it might yield significantly different results.
![inlet and outlet configurations](/ext/resources/images/2020/CFD-Figure-1.jpg)
In HVAC design, minimizing fluid recirculation as much possible ensures proper ventilation. In Figure 2, the velocity plot comparison obtained with CFD simulation shows that recirculation is reduced in Configuration 2 compared to Configuration 1. The temperature plot illustrates the significance of this reduced recirculation.
![velocity](/ext/resources/images/2020/CFD-Figure-2.jpg)
![Temperature distribution](/ext/resources/images/2020/CFD-Figure-3.jpg)
The temperature plot in Figure 3 shows the domain in Configuration 2 has a lower temperature value than the corresponding points in Configuration 1. Additionally, the average temperature at the center section of Configuration 1 is 303.9°K, whereas the second case has a lower average temperature at 301.8°K.
For data center cooling, it is crucial to understand the airflow. This information helps create a channel for effectively removing the heat dissipated from server vents. The illustrations compare two basic configurations. In general, there can be many other possibilities thatachieve even better results. Temperature distribution and pressure differences should be uniform to maintain the conditions inside the room.
Conclusion
Cloud-based solutions for CAE are posing a new paradigm for high-performance computing. Their goal is to make simulation technology accessible to engineers and allow them to, ultimately, create better products, faster.