Temperature management in data centers

El-Sayed, Nosayba; Stefanovici, Ioan; Amvrosiadis, George; Hwang, Andy; Schroeder, Bianca

doi:10.1145/2254756.2254778

Cited by 120 publications

(47 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…the power tends to increase linearly with temperature (as reported by [9]); (ii) the exponential model presented in Equation (1) tends to over-estimate the failure rate increase with temperature.…”

Section: Server Lifetime Modelmentioning

confidence: 89%

“…More in depth, in [9] authors report temperature measurements of disks, memories, and components in DCs, showing that the largest impact on failure rates is due to the variation of temperature rather than its increase. Moreover, the authors show that the failure rate of HW components is linear with temperature rather than exponential.…”

Section: Related Workmentioning

confidence: 99%

“…However, we stress that Equation (1) is just a first-order approximation. For example, according to [9], failure rates exhibit a linear dependency with temperature (rather than exponential). Therefore, we will consider this aspect when defining a model for the server failure rate.…”

Section: Physical Factors Affecting Server Lifetimementioning

confidence: 99%

See 2 more Smart Citations

Lifetime-Aware Cloud Data Centers: Models and Performance Evaluation

et al. 2016

View full text Add to dashboard Cite

We present a model to evaluate the server lifetime in cloud data centers (DCs). In particular, when the server power level is decreased, the failure rate tends to be reduced as a consequence of the limited number of components powered on. However, the variation between the different power states triggers a failure rate increase. We therefore consider these two effects in a server lifetime model, subject to an energy-aware management policy. We then evaluate our model in a realistic case study. Our results show that the impact on the server lifetime is far from negligible. As a consequence, we argue that a lifetime-aware approach should be pursued to decide how and when to apply a power state change to a server.

show abstract

Section: Server Lifetime Modelmentioning

confidence: 89%

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Lifetime-Aware Cloud Data Centers: Models and Performance Evaluation

et al. 2016

View full text Add to dashboard Cite

show abstract

“…A recent work [19] reports that a high variability in temperature has a stronger effect on hardware reliability. Based on their observations, the probability of hardware failure is almost double when the CoV (coefficient of variation) in temperature is bigger than 0.0074.…”

Section: Pulsation Thermal Attackmentioning

confidence: 99%

“…• C can save approximately 2-5 percent of the cooling power [19]. Thus, there is a trend in data centers to raise the highest set temperature from 75…”

Section: Introductionmentioning

confidence: 99%

Reduced Cooling Redundancy: A New Security Vulnerability in a Hot Data Center

Gao

Wang

et al. 2018

Proceedings 2018 Network and Distributed System Security Symposium

View full text Add to dashboard Cite

Abstract-Data centers have been growing rapidly in recent years to meet the surging demand of cloud services. However, the expanding scale and powerful servers generate a great amount of heat, resulting in significant cooling costs. A trend in modern data centers is to raise the temperature and maintain all servers in a relatively hot environment. While this can save on cooling costs given benign workloads running in servers, the hot environment increases the risk of cooling failure. In this paper, we unveil a new vulnerability of existing data centers with aggressive cooling energy saving policies. Such a vulnerability might be exploited to launch thermal attacks that could severely worsen the thermal conditions in a data center. Specifically, we conduct thermal measurements and uncover effective thermal attack vectors at the server, rack, and data center levels. We also present damage assessments of thermal attacks. Our results demonstrate that thermal attacks can (1) largely increase the temperature of victim servers degrading their performance and reliability, (2) negatively impact on thermal conditions of neighboring servers causing local hotspots, (3) raise the cooling cost, and (4) even lead to cooling failures. Finally, we propose effective defenses to prevent thermal attacks from becoming a serious security threat to data centers.

show abstract