Performance-aware thermal management via task scheduling

Zhou, Xiuyi; Yang, Jun; Chrobák, Marek; Zhang, Youtao

doi:10.1145/1736065.1736070

Cited by 25 publications

(11 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Speedup of multicore processors under thermal constraints is determined in [27]. An OS-level technique for job scheduling is proposed in [28]. Thermal aging is addressed in [29].…”

Section: B Power and Thermal Managementmentioning

confidence: 99%

TheSPoT: Thermal Stress-Aware Power and Temperature Management for Multiprocessor Systems-on-Chip

Iranfar

Kamal

Afzali-Kusha

et al. 2018

IEEE Trans. Comput.-Aided Des. Integr. Circuits Syst.

View full text Add to dashboard Cite

Abstract-Thermal stress including temperature gradients in time and space, as well as thermal cycling, influences lifetime reliability and performance of modern Multiprocessor Systemson-Chip (MPSoCs). Conventional power and temperature management techniques considering the peak temperature/power consumption do not provide a comprehensive solution to avoid high spatial and temporal thermal variations. This work presents TheSPoT, a novel multi-level thermal stress-aware power and thermal management approach for MPSoCs. At the top level, core consolidation and deconsolidation is performed based on peak temperature, thermal stress, and power consumption constraints. These constraints are also used at the next level, where operating frequencies are determined. At this level we obtain optimal core frequencies by solving a convex optimization problem. However, thereafter, to reduce the runtime overhead in large MPSoCs, we alternatively propose to use a fast heuristic algorithm. The efficacy of the proposed approaches in reducing the thermal cycles and temporal/spatial temperature gradients is evaluated by comparing the results with the state-of-the-art methods. The evaluation performed on 4-core, 8-core, and 16-core MPSoCs, using PARSEC benchmarks, reveals a considerable reduction in thermal stress. For the 8-core MPSoC case study, on average, for the proposed heuristic(optimal) approach, the mean time to failure improved by 47(35) % compared to the state-of-the-art techniques with only 6(4) % performance degradation. Also, our simulations show that TheSPoT is more efficient in thermal stress reduction when more heterogeneous workloads are used.

show abstract

“…Speedup of multicore processors under thermal constraints is determined in [27]. An OS-level technique for job scheduling is proposed in [28]. Thermal aging is addressed in [29].…”

Section: B Power and Thermal Managementmentioning

confidence: 99%

TheSPoT: Thermal Stress-Aware Power and Temperature Management for Multiprocessor Systems-on-Chip

Iranfar

Kamal

Afzali-Kusha

et al. 2018

IEEE Trans. Comput.-Aided Des. Integr. Circuits Syst.

View full text Add to dashboard Cite

show abstract

“…4 Xi'an Satellite Tracking, Telemetering and Control Center, Xi'an, China. 5 Inner Mongolia Agricultural University, Hohhot, China.…”

Section: Acknowledgementsmentioning

confidence: 99%

“…There are a number of studies separately dealing with energy minimization in data centers. For example, [2][3][4][5][6] studied IT energy consumption, others [7][8][9] inclined to optimize cooling energy consumption. However, the aforementioned studies only concentrated on the energy consumption of a subset of system components and overlooked the interactions among them.…”

Section: Introductionmentioning

confidence: 99%

Dynamic thermal and IT resource management strategies for data center energy minimization

Wan

Yang³

et al. 2017

J Cloud Comp

View full text Add to dashboard Cite

Energy-hungry data centers attract researchers' attention to energy consumption minimization-a challenge confronted by both industry and academia. To assure the server reliability, its instantaneous temperature is generally controlled within a preset threshold. Nonetheless, field studies indicated that the occasional violation of extreme temperature constraints hardly affect the system reliability in practice. Therefore, strictly restraining the server temperature may contribute to meaningless energy consumption. As a response to this limitation, this paper presents a dynamic control algorithm without violating the average temperature constraint. We formulate a "soft" Server Temperature-Constrained Energy Minimization (STCEM) problem, where the object function consists of IT and Heat Ventilation & Air Conditioner (HVAC) energy. Based on the Lyapunov optimization, two algorithms, i.e., linear and quadratic control policies, are proposed to approximately solve the STCEM problem. The non-negative weight parameter V is introduced to trade off the energy consumption against server temperature constraint. Furthermore, extensive simulations have been carried out to evaluate the system performance for the proposed controlled algorithms. The experimental results demonstrate that the quadratic control policy outperforms the linear counterpart on the STCEM problem. Specifically, the energy consumption and server temperature constraint are well balanced when V ≈ 5000.

show abstract

“…Thermal-aware scheduling at both the chip level and the data center level has been proposed in some works [88,89,111,126]. The idea is to decrease the heat generated by microchips and servers, respectively, through workload scheduling.…”

Section: Thermal-aware Schedulingmentioning

confidence: 99%

Sustainable Cloud Computing

Domdouzis

2015

Green Information Technology

View full text Add to dashboard Cite

Energy consumption of the data centers worldwide is rapidly growing fueled by everincreasing demand for Cloud computing applications ranging from social networking to e-commerce. Understandably, ensuring energy-efficiency and sustainability of Cloud data centers without compromising performance is important for both economic and environmental reasons. In order to achieve these objectives, this dissertation develops a cyber-physical multi-tier server and workload management architecture which operates at the local and the global (geo-distributed) data center level.This dissertation devises optimization frameworks for each tier to optimize energy consumption, energy cost and carbon footprint of the data centers. The proposed solutions are aware of various energy management tradeoffs that manifest due to the cyber-physical interactions in data centers, while providing provable guarantee on the solutions' computation efficiency and energy/cost efficiency. The local data center level energy management takes into account the impact of server consolidation on the cooling energy, avoids cooling-computing power tradeoff, and optimizes the total energy (computing and cooling energy) considering the data centers' technology trends (servers' power proportionality and the cooling power efficiency). The global data center level cost management explores the diversity of the data centers to minimize the utility cost while satisfying the carbon cap requirement of the Cloud and while dealing with the adversity of the prediction error on the data center parameters. Finally, the synergy of the local and the global data center energy and cost optimization is shown to help towards achieving carbon neutrality (net-zero) in a cost efficient manner.

show abstract

Performance-aware thermal management via task scheduling

Cited by 25 publications

References 46 publications

TheSPoT: Thermal Stress-Aware Power and Temperature Management for Multiprocessor Systems-on-Chip

TheSPoT: Thermal Stress-Aware Power and Temperature Management for Multiprocessor Systems-on-Chip

Dynamic thermal and IT resource management strategies for data center energy minimization

Sustainable Cloud Computing

Contact Info

Product

Resources

About