Masteruppsats

Energieffektivitet i datahallar

Sammanfattning

The number of data centers and their capacity is increasing throughout the world. The

energy used in these data centers and the emissions from the use are increasing at the

same rate. The total use of energy will double from 2008 to 2011 based on the trends in

data centers use. Research in this area says that the emissions from data centers will

have surpassed the level of airline traffic in 2020. The index used today to compare how

efficient a data center is, called PUE, is lacking some information. It is not able to

compare different kinds of cooling solutions and it does not take in account the large

amount of heat produced in a data center. This development and the limitations of the

existing index calls for an improved way of measuring energy efficiency in data centers.

This thesis suggests a changed index, which can compare different cooling solutions

such as district cooling compared to chillers. The suggested index also gives a better

number to facilities that re-use the heat energy produced by the servers and the chillers.

During this work the new index has been called NPUE, Net Power Usage Effectiveness.

This name might be changed before an introduction to users and the market.

The major changes from PUE to NPUE can be summed up as:

?

NPUE measures the net energy flow to and from the data center, where PUE

measures the electric energy or power delivered to the data center.

?

NPUE measures the energy used over a period of 12 months.

The net energy to and from the data center incorporates energy in other forms than

electric energy, such as cooling in the form of district cooling to the data center and heat

energy delivered from the data center.

NPUE is supplemented by a few more data about the facility. These are data of

emissions, in the form of CO

2, NOxand SOx, from the energy used in the data center. It

includes a comparison of the security of the infrastructure in the data center based on

Tier-levels. Finally a comparison of how efficient the data center is using the servers

Läs hela uppsatsen hos lärosätet