'Data center' simulation project by sjesu_rajendra


#1

I created a new simulation project called 'Data center':

Validation for data center - by Gao et al


More of my public projects can be found here.


#2

@sjesu_rajendra - awesome project! Did some addition post-processing:


#3


#4

Description


 
Data centers are critical, energy-hungry infrastructures that operate round the clock. They provide computing functions that are vital to the daily operations of top economic, scientific, and technological organizations around the world. The amount of energy consumed by these centers is estimated at 3% of the total worldwide electricity use, with an annual growth rate of 4.4%. Naturally, this has a tremendous economic, environmental, and performance impact that makes the energy efficiency of cooling systems one of the primary concerns for data center designers, ahead of the traditional considerations of availability and security. [1]
 
Studies also show that the largest energy consumer in a typical data center is the cooling and ventilation infrastructure (50%), followed by servers and storage devices (26%) [2]. Thus, in order to control costs while meeting the increasing demand for data center facilities, designers must make the cooling infrastructure and its energy efficiency their primary focus.
 

Project Goals


 
The current project involves conducting a Computational Fluid Dynamics (CFD) analysis of the airflow inside the data center with the help of the SimScale platform. With the end goal of studying the airflow patterns and designing the ventilation system, we simulate a sample meshed geometry of the data center along with preset boundary conditions.
 
Our results will reveal a deeper insight into the further steps needed to be taken to make design alterations or the data center and/or the ventilation system.
 

Geometry


 
The geometry is constructed such as that of any standard data centre. The data center under consideration, houses several physical servers lined up together in the form of racks. A more accurate representation of the geometry can be found in the images below.
 
image
 
Isometric view
 
image
 
Internal view
 

Meshes


 
A hex-dominant parametric (only CFD) mesh is generated and mapped onto the imported geometry. The meshing operation requires 16 computing cores and the resultant mesh contains over 5 million nodes. An overall and a fine mesh refinement is made on the surface of the entire body in order to have a higher level of accuracy in our results.
 
image
 
Isometric view of meshed geometry
 
image
 
Internal view of meshed data center geometry
 

Simulation


 
Post meshing, we run a Steady State simulation, with an analysis type of Convective Heat Transfer. The turbulence model chosen for this simulation is k-omega SST due to its good behaviour in adverse pressure gradients and separating flow.
 
The simulation is set up with the following boundary conditions:
 

  • 4 Inlets: Velocity inlets with a fixed volumetric flow rate
  • Outlet: Pressure outlet at atmospheric pressure
  • Heat Source: As heat is released from the vents of the servers and storage devices, they are set to be the heat source in the current simulation.
  • Walls: All the remaining surfaces are assigned as walls, with no slip and zero temperature gradient.
     

Results and Conclusion


Below, we can see the results of the simulation post-processing.
 


 
The image above shows the top view of the velocity contour at the mid-section of the room. The velocity is highest in the proximity of the server stacks, which are the heat source. Due to the high-temperature region created, there is a lower pressure, resulting in an increased flow of air into that region. Effectively, this air gets heated up and is carried away into the low-temperature regions, situated outside the stack area.
 
The figure below shows a similar view of the temperature contour. As can be observed, the high-temperature regions near the servers plummet to lower temperature due to the cross-flow across the room. Thus, we may conclude that ventilation is effective to provide sufficient cooling.
 

 
The figure below depicts the vector flow of the air as seen from a vertical cross-section.
 

 

References


 
[1] W. V. Heddeghem et al., Trends in worldwide ICT electricity consumption from 2007 to 2012, Comput. Commun., vol. 50, pp. 64–76, Sep. 2014
 
[2] Top 10 energy-saving tips for a greener data center, Info-Tech Research Group, London, ON, Canada, Apr. 2010, http://static.infotech.com/downloads/samples/070411_premium_oo_greendc_top_10.pdf