'Data Center Cooling Cost Reduction - Editorial Demo' simulation project by vaibhav_s


#1

I created a new simulation project called 'Data Center Cooling Cost Reduction - Editorial Demo':

The aim of this project is to showcase how SimScale Platform could be used to perform CFD Analysis to optimize cooling cost of a Data Center. From the CFD results, Rack Cooling Index (RCI) is calculated and cooling cost optimized.


More of my public projects can be found here.


#2

Data Center Cooling Cost Reduction with CFD



Introduction

Reducing Data centers energy consumption is one of the major problems that engineers and researchers will have to solve in the next years. Indeed the required energy has double over the past past 5 years and is supposed to do so in the next 5 years, reaching a consumption of more than 1000 billion kWh.

On a “thermal” point of view, a Data center aim is to provide an appropriate working environment to its Information Technology (IT) material. Data centers are considered as mission critical facilities which means that the cooling system must comply with the server’s temperature and hygrometry requirement. Actually a server working at too high temperature increases the risks of experimenting failure, rise in power consumption and even irreversible damage. On the opposite a too low cooling temperature might lead to Electrostatic Discharge (ESD)or appearance of water condensation.

Data centers that use perforated tiles, placed on the floor and in front of the server racks, to deliver cold air typically have a raised floor with an under-floor plenum. The data center server rack layout is predominantly in the hot-aisle/cold-aisle configuration. This is meant to separate the hot exhaust air from the cold inlet air. There are a complex airflow pattern and distribution in the under-floor plenum resulting from the placement of the perforated tiles and CRAC units.

The exact prediction of flow distribution is difficult. CFD analysis can help analyze the turbulent flow conditions, complexity of geometry, flow paths and blockages in the plenum, for example, from cabling and chilled water distribution pipes.

Data Center Thermal Performance

Rack cooling index (RCI)
The rack cooling index (RCI) is a measure of how well the inlet temperatures conform to a selected inlet temperature standard, such as ASHRAE. There are two parts to it, high end (HI) and low end (LO).

RCI- HI is a measure of the equipment health at the high end of the temperature range. Over-temperature conditions exist when one or more equipment intake temperatures exceed the max recommended temperature.

The RCI-LO is a complement to RCI-HI especially when the supply condition is below the minimum recommended temperature.

Facilities should be designed and operated to target the recommended range. Electronic equipment should be designed to operate within the extremes of the allowable operating environment. Prolonged
exposure to temperatures outside the recommended range can result in decreased equipment reliability and longevity.

ASHRAE Standards - “The recommended equipment intake temperature range is 20-25°C and the allowable range is 15-32°C”

Project Overview

The aim of the project is to optimize the rack cooling effectiveness and reduce the cooling cost of a typical data center design. Two key design parameters are evaluated :-
(1) Supply Temperature (2) Supply Airflow Rate

These parameters impact both the rack cooling effectiveness and energy cost for cooling the space. With CFD. we analyze and calculate the flow conditions such as intake temperatures for the server racks which is the most critical parameter for cooling effectiveness.

Different combinations of supply temperature and supply airflow rate are analyzed for their impact on the cooling effectiveness and cooling cost.

The modeled Data center consists of 4 rows of 13 server racks each for a total of 52 server racks. Each of them dissipate 4 KW of heat. As a result, the total space load is 208 KW. Total rack airflow is 56680 m^3/hr.
The Data center houses 4 CRAC units.

Supply temperature is varied from 13 - 21 Degree Celsius and supply airflow rate is varied from 80% - 140% of total rack airflow rate to find the best combination of the two parameters to optimize cooling effectiveness and cooling cost.

CAD Model

CADlabels1CADlabels2

Mesh

Simulation Results

Server Intake Temperatures

Server Temperatures


Velocity Distribution

Pressure Drop at Perforated Plates and Server Grills

Cost Functions for Chiller and Fan

RCI Results & Net Energy Savings


#3

Are those tiles perforated?