Data center design and infrastructure final exam

Choose the best answer to each question then click the "Check Score" button at the bottom of the page to view your score. Incorrect responses will be displayed red. View explanations and references to each question by following the "Check answer" link. Good luck!

1. According to a research study by the Princeton, N.J.-based site selection specialists, The Boyd Company Inc., there are three major factors affecting the decision of where to build a data center. Based on these factors, which region in the U.S. seems to be the most logical place to build a data center?

a)The east and west coasts-Despite higher costs in other areas, the talent pool justifies the expense.
b)The northen midwestern states-Power is cheaper and natural disasters are fewer.
c)The south-Cheap labor and cheap real estate trumps the threat of natural disasters.
d)Pacific noncontiguous (Alaska and Hawaii)-It's where power is the cheapest, right?

Check answer

2. Sun's Rob Snevely talks about rack location units (RLU) in data center design. Why are these important?

a)RLUs are part of a data center monitoring software that assigns locations to server racks, making it easier to monitor them.
b)RLUs are devices attached to server racks by the facilities department that help them keep an estimate of how many servers are being brought into the data center.
c)RLUs are a way to describe any particular location on the floor -- including requirements for power, cooling, bandwidth, weight -- so you can do capacity planning.
d)They are not important to data center design.

Check answer

3. HP is planning to consolidate 85 worldwide data center facilities down to six, taking HP's data center real estate down from 500,000 square feet to 300,000. What is cited as a major, but often unsung, component to their consolidation strategy?

a)Application consolidation -- it's a huge cost saver in licensing, management resources and hardware support.
b)Losing the raised floor and going with a high-density approach to computing and cooling.
c)Decreasing the size of the staff -- this will force the IT department to prioritize projects and work smarter, not harder.
d)Architecture diversification -- This allows IT to optimize applications and weed-out unnecessary programs.

Check answer

4. According to founder and executive director of The Uptime Institute Ken Brill, sealing the cable openings at the back of racks and taking perforated tiles in the hot aisle and putting them in the cold aisle are two strategies that do what in a typical data center?

a)Doing this increases bypass airflow, which results in better airflow management.
b)Doing this enables data center staff to isolate hot spots, so more cooling units can be brought in to address the problem.
c)Doing this increases that data center temperature and is not recommended -- these are strategies that have been tested, but have had negative results.
d)Doing this reduces bypass airflow, which results in getting the air to where it needs to go.

Check answer

5.Though backup power from flywheel technology is not a new concept, UPS providers are increasingly becoming interested using flywheels either as an alternative to or in tandem with the traditional battery UPS. Why?

a)Batteries' shortcomings are due to their limited lifetime -- using a flywheel to reduce battery cycles can extend the overall life of the backup system.
b)This is not really true -- flywheels depend on the primary source to begin spinning, rendering them less efficient because of this initial power consumption.
c)The mechanical nature of flywheels makes them a favorable option to the chemical processes of batteries.
d)This is not really true -- flywheels don't provide continuous power to make them useful at this point.

Check answer

6.Energy efficiency is not only socially more responsible, it's ecnomomically mandatory. According to APC's Neil Rasmussen, what areas of the data center have the biggest margin for improvement?

a)Servers -- Expanding EPA's the Energy Star program to include servers will have the most dramatic effect on energy efficiency in the data center.
b)Data center design -- IT departments tend to over-size and count on their companies expanding to capacity; in the meantime, they're inefficient.
c)CRAC units -- We need to more past forced-air cooling and use water because water is a more efficient medium for removing heat.
d)If data center staff would just turn the light off when they're not in the data center, they would have a huge impact on efficiency.

Check answer

7.True or False -- The following statement accurately describes the basic procedure for measuring server energy consumption of 1 and 2U machines as currently put forth by server industry participants:


Server frequency, voltage, power factor and total harmonic distortion are measured and compared to CPU utilization. A curve with power output on the y-axis and percentage of workload on the x-axis is created. Measurements for workload are in 10% increments from 0% to 100%, allowing data center managers to examine the curve depending on how busy they expect their servers to be.


Check answer

8. In the search for more efficient ways to use available power, some companies are exploring, and even implementing, direct current (DC) as opposed to alternating current (AC) systems. However, there are pros and cons to both approaches. Which of the following statements are true concerning this debate?

a)AC-powered systems are more efficient because they only require the power to be converted once, avoiding energy loss.
b)DC power is easier to move over long distances, making it a more practical option.
c)Both a and b are true.
d)Neither a or is true.

Check answer

9. Hydrogen fuel cells can provide clean and reliable energy without some of the limitations of more traditional power supplies. But there are concerns about certain aspects of hydrogen fuel cell technology. Which of the following statements is true about hydrogen fuel cells?

a)The upfront cost of purchasing and installing a PAFC today is much higher than the cost of a diesel generator or gas turbine, but the 20-year payback is comparable due to the fuel cell's lower maintenance and better efficiency.
b)Hydrogen fuel cells as a backup power system may be favorable option when conventional systems aren't permitted, such as an alternative to a diesel generator because of structural, noise, planning or emissions restrictions.
c)Hydrogen fuel cell power is currently impractical according to some experts because hydrogen is still manufactured from natural gas -- a dirty process -- and using electricity to separate the hydrogen is less efficient than simply using electricity in the first place.
d)All of the above are true.

Check answer

10.According to data center power and cooling expert Robert Macfarlane, what is the biggest problem in determing what size of UPS to go with?

a)Most manufacturers do not provide adequate power information on their equipment, making it difficult to determine actual load.
b)Equipments' power factor (pf), or ratio of reactance power versus actual available power, is rarely stated on equipment.
a)The lack of continuity in power factor (pf) rating between American and European manufacturers is the biggest problem -- in the U.S., it is assumed to be between 1.0 and .8 versus the European rating of .95 to .98.
a)Determing what vendor to purchase equipment from is always the most difficult part.

Check answer

11.Air side economizers, which bring in cooler outside air to lower that data center's temperature, have the potential to save companies a lot of money on cooling. But there is also a lot of skepticism about the viability of air side economizers from people like cooling expert Steve Madara and ASHRAE member Don Beaty. Which of these concerns have been cited?

a)Contaminants like pollen or construction dust.
b)Conditioning the outside air to the appropriate humidity level potentially negates the energy saved by using air side economizers.
c)Air side economizers are only practical for specific geographies.
d)All of the above.

Check answer

12.Raised floor cooling has been the standard for many years. But during the late 1990s, many companies started experimenting with overhead cooling, claiming that handling IT equipment was easier if the cabling was overhead, and the influx of smaller servers made water piping unnecessary. In an IBM study, researchers compared the methods. What were the findings?

a)Both cooling strategies provided similar amounts of cooling with irrelevant levels of variance.
b)With chillers operating at 100% and 80% levels, overhead cooling resulted in cooler rack intake temperatures.
c)With chillers operating at 100% and 80% levels, raised-floor cooling resulted in cooler rack intake temperatures.
d)The results were largely inconclusive.

Check answer

13.Additional cooling units aren't always the answer to mitigating hot spots. Air delivery has everything to do with the ability of ACs to bring down the temperature in a data center. According to Bob Macfarlane, there are three primary causes of poor air delivery: Bypass air, air leakage, and ______. According to Syska Hennessy researchers Christopher Johnson and Vali Sorell, ___ modeling may assist data center designers in locating these air delivery failures.

a)overhead cooling, UPS
b)under-floor blockage, CFD
c)filtration, PF
a)high-density servers, CFM

Check answer

14.Using Bob Macfarlanes formula for determining cubic feet per minute (CFM), how many CFMs will be required to cool the following configuration (do not include correction factor):

  • 10 blades at 300 Watts each
  • Entering air is 50 degrees
  • Exiting air is 80 degrees

    a)100 CFM
    b)265 CFM
    c)340 CFM
    d)none of the above.

    Check answer

15.There are many facets to the liquid versus air cooling debate. On one hand, ASHRAE TC 9.9's Don Beaty says that because water is 3,500 times more efficient than air, we'll have to make the move to water to maximize cooling efficiency. On the other, APC's Neil Rasmussen says that the dynamic nature of enterprise data centers makes it impractical to use build a liquid cooling infrastructure. But even within the liquid cooling camps, there are different approaches. Which of the following statments describes IBM's Cool Blue option?

a)Liquid is delivered directly to the rack. Once the liquid is within the vicinity of the rack, it is then used to cool air locally, which in turn cools the equipment.
b)Pumps in an inert liquid and sprays it directly onto the hot chip to cool it down.
c)Enables server racks to consume up to three times as much power as its existing racks -- up to a staggering 30 kilowatts -- while keeping temperatures at safe operating levels.
d)Removes up to 55,000 BTUS of heat from fully populated server racks and potentially lowers energy costs as much as 15 percent.

Check answer

This was first published in March 2007

There are Comments. Add yours.

TIP: Want to include a code block in your comment? Use <pre> or <code> tags around the desired text. Ex: <code>insert code</code>

REGISTER or login:

Forgot Password?
By submitting you agree to receive email from TechTarget and its partners. If you reside outside of the United States, you consent to having your personal data transferred to and processed in the United States. Privacy
Sort by: OldestNewest

Forgot Password?

No problem! Submit your e-mail address below. We'll send you an email containing your password.

Your password has been sent to: