ASHRAE Journal - November 2014 - 83

COLUMN DATA CENTERS
Donald L. Beaty

X-Factor Explained
With the global push on energy-efficient operation, the third edition of ASHRAE's
Thermal Guidelines for Data Processing Environments sought to provide more actionable
information that would empower data center owners and operators to make businesscentric decisions when it came to the operation of their data center facilities.

The term "x-factor" refers to a dimensionless value of
relative server failure rate vs. ambient inlet temperature. Since the x-factors are relative, there needed to be
a baseline. It was determined that the most appropriate
baseline would be to use a common data center operating temperature of 68°F (20°C) at the inlet of the IT
equipment. This condition would represent an x-factor
value of 1.00.
Of course, there are many types of IT equipment and
each of them may have a different failure rate profile,
but the x-factors are based on the most common IT
equipment in data centers: volume servers. ASHRAE
TC 9.9's Datacom Equipment Power Trends and Cooling
Applications publication defines volume servers as "oneand two-processor socket systems typically packaged in
1U, 2U, pedestal, or blade form factors."
Another variable that needed to be considered was that
the failure rates are based on manufacturer reliability

temperature.

SERVER RELIABILITY TREND VS. AMBIENT TEMPERATURE - RELATIVE FAILURE RATES
DRY-BULB
TEMPERATURE

HARDWARE FAILURE RATE FOR VOLUME SERVERS (X-FACTOR)
AVERAGE

LOWER BOUND

UPPER BOUND

59°F

0.72

63.5 °F

0.87

0.80

0.95

68°F

1.00 (Baseline)

0.88

1.14

72.5°F

1.13

0.96

1.31

77 °F

1.24

1.04

1.43

81.5°F

1.34

1.12

1.54

86°F

1.42

1.19

1.63

95°F

1.55

1.35

1.74

104°F

1.66

1.51

1.81

113°F

1.76

1.67

1.84

Table reformatted by DLB Associates

X-Factor

TABLE 1 Relative ITE failure rate x-factor as a function of constant ITE air inlet

©ASHRAE

Energy-efficient data center operation through leveraging the expanded ASHRAE TC 9.9 allowable environmental envelopes was somewhat tempered by the
previously unknown impact that higher inlet temperatures would have on the failure rates of the IT equipment. To quantify this unknown, a conglomerate of the
major IT equipment manufacturers that make up the IT
OEM subcommittee of ASHRAE TC 9.9 determined and
published, for the first time, consensus relative failure
rates of IT equipment at different inlet temperature
conditions.
This groundbreaking information was included in the
third edition of Thermal Guidelines along with practical
examples of how to apply the information. This article
provides an overview of the relative failure rate information contained in the Thermal Guidelines publication.

TABLE 2 Temperature impact on annual server failure rate
VOLUME SERVER FAILURE RATE VERSUS AMBIENT TEMPERATURE

DRY-BULB
TEMPERATURE

AVERAGE RELATIVE
FAILURE RATE
(NOTES 2 AND 3)

ANNUAL SERVER FAILURE RATE
BASELINE
(NOTE 1)

ADJUSTED DUE TO
TEMPERATURE

D=B×C

59°F

0.72

3% Less Failures

68°F

1.00

Baseline

104°F

1.66

More
Failures

COMMENTS

1. Table uses the conservative upper bound of typical server failure rates of 2% to 4%.
2. Assumes continuous (7 × 24 × 365) operation with dry-bulb temperature at ITE inlet.
3. Based on years of server manufacturers' actual failures, component manufacturers' failure
data, and modeling (published by ASHRAE TC 9.9).

N OVEM BER 2014

ashrae.org

ASHRAE JOURNAL

ASHRAE Journal - November 2014

Table of Contents for the Digital Edition of ASHRAE Journal - November 2014