2. โข A data center is a large group of networked computer
servers typically used by organizations for the remote
storage, processing, or distribution of large amounts of
data
โข In 2007, 1.5% of the electricity consumption in the
United States was used for data centers
โข Of the 1.5%, one-third to one-half of this energy was used
for cooling
โข Data centers are most commonly cooled by air delivered
to electronic equipment from centralized cooling systems.
2
3. โข Data centers are one of the largest and fastest growing
consumers of electricity in the world
โข Data centers, which power the internet, are a key part
of the infrastructure in the United States
โข There has been a push in recent years to improve their
energy efficiency
โข The Villanova Steel Orca Research Center is a research
data center that will be built in Princeton, New Jersey
that will allow for testing of improved energy efficient
cooling strategies
3
4. 4
โข In 2013, US data centers
โ Consumed 91 billion kWh of electricity (2.5% of total consumption)
โ Emitted nearly 100 million metric tons of CO2 annually
โ Projected to increase to 140-200 billion kWh annually by 2020
โข Data centers are commonly cooled by air from centralized cooling systems
โ 40% - 50% of energy consumed in data centers is for cooling
5. 5
Optimize cooling system design of a research data center in
both load capacity and thermal efficiency using CFD
Villanova Steel Orca Research Center (in design stages) will allow for
investigation of:
โข Cooling strategies- Perimeter, In-row, Overhead, or Hybrid cooling
โข Containment- Hot & cold containment
โข Layout & load distribution
6. 6
Determine optimal operating conditions & design layout while
meeting thermal constraints
โข Minimization function: Total energy consumed for cooling
โข Design variables:
โข Air conditioning units (ACU) air flow rate
โข Chiller supply temperature setpoint (CHWST)
โข Constraint: Racksโ maximum inlet temperature < 85หF
โข Investigate effects of containment
7. โข This research is motivated by a push for improved
energy efficiency for data centers
โข State of the art data center cooling strategies will
be investigated
โข CFD software 6SigmaRoom provides ability to
easily test various strategies such as high density
zones and hot and cold aisle containment
โข Efficient strategies will be validated at VSORC
facility
7
8. 8
Enclosures to control travel path of airflow
CRAH
Containments:
โข Enclose hot or cold aisles
โข Prevent premature mixing of hot & cold streams โ reduce cooling load
โข Requires racks layout & height uniformity
โข Are difficult to install and remove
Both hot & cold containment strategies
were investigated
9. โข Containment enclosures are installed in data
centers to control the path of airflow travel
โข Containment helps in reducing the air inlet
temperatures to the IT equipment
โข Containment does not change air movement, it
can be difficult to install and remove, and it
requires uniformity in the server cabinet layout
and height
9
10. โข Addition of containment (hot or cold) does not
directly alter power usage effectiveness (PUE), a
popular measure of the energy efficiency of a
data center:
โข Containment reduces premature mixing of hot
and cold air streams, resulting in reduced cooling
load
10
PUE =
๐๐๐ก๐๐ ๐น๐๐๐๐๐๐ก๐ฆ ๐๐๐ค๐๐
๐ผ๐ ๐ธ๐๐ข๐๐๐๐๐๐ก ๐๐๐ค๐๐
11. The Center for Energy-Smart Electronic Systems
โข Containment prevents the mixing of different airflow
paths and damage to the IT equipment
โข Containment can help reduce the inlet temperatures of
the IT equipment and the needed cooling power
Without Hot Aisle Containment Full Hot Aisle Containment
12. 12
High power server/racks (high kW)
are clustered together in one zone
โ Allows for higher computing
power per unit area
High density zones cause cooling
issues requiring additional cooling
Additional cooling accomplished
through In-row coolers
high density zones in-row
14. 1. Two distinct VSORC design configurations
have been modeled
๏ Each model required extensive clean up of initial
layout by removing numerous collision and cooling
errors
2. Optimization model
๏Design variables: total supply flow rate and chiller
supply temperature
๏Design constraint: maximum rack inlet temperature
no greater than 85ยฐF
14
15. 3. The baseline model must be used to create
different containment configurations
4. Perform CFD simulations to find most efficient
containment configuration
5. Create matrix based on power, max inlet
temperature, and total supply flow rate
6. Utilize Minitabยฉ to create predictive equations
and identify optimal operating point
7. Based on results, either validate or reoptimize
15
17. 17
Without Hot Aisle Containment
Full Hot Aisle Containment
Hot aisle containment
Full
Partial*
None
Cold aisle containment
Full
Partial*
None
6 Models
โข Industryโs best practices employed to create models
Containment leakage
Equipment gaps & interferences
Supply tile locations
Chiller supply temperature setpoint range
Total flow rate range
* Partial configurations enclosed only high density zones
18. 18
IT Equipment
8 rows
36 racks
550 kW total IT power (total heat load)
2 rows of 10kW racks
2 rows of 5kW racks
4 rows of high density 40kW racks
Cooling Equipment
2 CRAH units
16 In-Row coolers
Containment
Various containment strategies
hot aisle
cold aisle cold aisle
hot aisle
cold aisle
hot aisle
CRAHCRAH
hot aisle
21. 21
Flow rate effect on max inlet temperature
โ
CHWST effect on max inlet temperature
โ
โ
22. 22
๐๐๐๐ 2
๐๐๐๐ 1
=
๐2
๐1
3
Tsupply = chiller supply temperature (CHWST)
COP = chiller coefficient of performance
๐ = total heat load
Fan power consumption is calculated from the use of the Fan Affinity Laws
๐๐๐๐ = fan power ; ๐ = total supply flow rate
Chiller power consumption is calculated from a correlation developed by HP
๐ถ๐๐ = 0.0068 (๐๐ ๐ข๐๐๐๐ฆ)2
+ 0.0008 (๐๐ ๐ข๐๐๐๐ฆ) + 0.458
๐๐โ๐๐๐๐๐ =
๐
๐ถ๐๐
(๐ป๐๐๐๐ ๐๐๐๐๐) ๐ท ๐๐๐๐๐= ๐ท ๐๐๐ + ๐ท ๐๐๐๐๐๐๐
23. 23
Regression analysis employed to obtain predictive equations for:
Total system power consumption ๐๐ก๐๐ก๐๐= 1609 + 0.00223 ๐๐ก๐๐ก๐๐ โ 22.913๐๐ ๐ข๐๐๐๐ฆ
Maximum server inlet temp. ๐ ๐๐๐ฅ= 80.076 โ 0.000681 ๐๐ก๐๐ก๐๐ + 0.99656๐๐ ๐ข๐๐๐๐ฆ
(r2 =94.17% & 99.89%)
๐ = total supply flow rate
Tsupply = CHWST
24. 24
โข Both chiller supply temperature setpoint & ACU total supply
flow rate impact data center total power consumption
โข CHWST has much greater effects on data center thermal
efficiency
Parameter Value
Total Supply Flow Rate, cfm 95,200
CHWST, ยฐF 70
Total Power Consumption, kW 218
Maximum Inlet Temperature, ยฐF 85
25. โข The supply temperature setpoint of the chiller was found
to significantly change the power consumption of the data
center
โข The total supply flow rate was found to only slightly
change the power consumption of the data center
โข A combined CFD & DOE (w/factorial analysis)
methodology was developed for design & optimization of
a data center
โข 1st law analysis was employed to determine cooling
systemโs optimal setting for improved energy efficiency
25
26. โข Containment is beneficial in reducing air inlet
temperatures, leading to an improvement in
energy efficiency
โข CFD programs like 6SigmaRoom can be
effectively used to identify optimal operating
points of total supply flow rate and chilled water
supply temperature setpoint for a data center
โข Hybrid cooling is effective in handling high
density IT zones of a data center
26
27. โข My research advisors, Dr. Kamran Fouladi and Dr.
Aaron Wemhoff of the Mechanical Engineering
Department along with Dr. Joseph Pigeon from the
Mathematics and Statistics Department
โข Thomas Wu and Aitor Zabalegui from Future
Facilites Inc.
โข The National Science Foundation (NSF) Research
Experience for Undergraduates (REU)
โข NSF I/UCRC in Energy-Smart Electronic Systems