Continuous Frequency Adjustment Technique Based on Dynamic Workload Prediction

Jung, Hwisung; Pedram, Massoud

doi:10.1109/vlsi.2008.98

Cited by 18 publications

(21 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The multinomial logistic regression model works at run-time and is used to classify workloads with uncertainty for energy minimization. The maximum likelihood estimation works on training set to 1 The derivation steps are omitted for space limitation. determine the parameters of the model in order to improve the efficiency of the probabilistic classification.…”

Section: B Maximum Likelihood Estimationmentioning

confidence: 99%

“…RESULTS The proposed run-time approach is validated on Texas Instrument's PandaBoard featuring ARM A9 cores and Intel quad-core system running Linux. The proposed approach is compared with one representative approach from each category of related works -the reinforcement learning-based technique of [4], the prediction-based DVFS technique of [1] and the multinomial logistic regression-based technique of [9].…”

Section: B Parameter Fixingmentioning

confidence: 99%

“…The average workload of the frames comprising a phase (inter) varies significantly across the different phases; however, the workload variation within each phase (intra) is relatively low. Proactive energy management involves predicting these dynamic workloads apriori to determine the most appropriate frequency for every phase such that performance constraint is satisfied while minimizing the energy consumption [1]. Studies have been conducted recently to use machine learning to determine the minimum frequency through continuous feedback from the hardware performance monitoring unit (PMU) [2]- [10].…”

Section: Introductionmentioning

confidence: 99%

“…RELATED WORKS Dynamic energy minimization has attracted significant attention both in industry and in academia [12]. Continuous frequency adjustment technique is proposed in [1] based on predicted workload, which is formulated as an initial value problem (IVP). The technique in [2]- [5] uses online learning to select the most appropriate frequency for the processing cores based on the workload characteristic of a given application.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Workload Uncertainty Characterization and Adaptive Frequency Scaling for Energy Minimization of Embedded Systems

Das¹,

Kumar²,

Veeravalli³

et al. 2015

Design, Automation &Amp; Test in Europe Conference &Amp; Exhibition (DATE), 2015

View full text Add to dashboard Cite

Abstract-A primary design optimization objective for multicore embedded systems is to minimize the energy consumption of applications while satisfying their performance requirement. A system-level approach to this problem is to scale the frequency of the processing cores based on the readings obtained from the hardware performance monitors. However, performance monitor readings contain uncertainty, which becomes prominent when applications are executed in a multicore environment. This uncertainty can be attributed to factors such as cache contention and DRAM access time, that are very difficult to predict dynamically. We demonstrate that such uncertainty can be controlled to make better decision on the processor frequency in order to minimize energy consumption. To achieve this, we propose a multinomial logistic regression model, which combines probabilistic interpretation with maximum likelihood (ML) estimation to classify an incoming workload, at run-time, into a finite set of classes. Every workload class corresponds to a frequency pre-determined using an appropriate training set and results in minimum energy consumption. The classifier incorporates (1) uncertainty with arbitrary probability distribution to estimate the actual frame workload; and (2) the frequency switching overhead, neither of which are considered in any of the existing approaches. The classified frequency is applied on the processing cores to execute the workload. The proposed approach is engineered into an embedded multicore system and is validated with a set of standard multimedia applications. Results demonstrate that the proposed approach minimizes energy consumption by an average 20% as compared to the existing techniques.

show abstract

Section: B Maximum Likelihood Estimationmentioning

confidence: 99%

Section: B Parameter Fixingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Workload Uncertainty Characterization and Adaptive Frequency Scaling for Energy Minimization of Embedded Systems

Das¹,

Kumar²,

Veeravalli³

et al. 2015

Design, Automation &Amp; Test in Europe Conference &Amp; Exhibition (DATE), 2015

View full text Add to dashboard Cite

show abstract

“…The higher the number of cores in the processor is, the more severe these issues become. Therefore, the ability of a DPM framework to scale well on a multicore processor by eliminating these overheads is becoming a critical requirement [3] [4].…”

Section: Introductionmentioning

confidence: 99%

Supervised Learning Based Power Management for Multicore Processors

Jung

Pedram

2010

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

Self Cite

107

View full text Add to dashboard Cite

-This paper presents a supervised learning based power management framework for a multi-processor system, where a power manager (PM) learns to predict the system performance state from some readily available input features (such as the occupancy state of a global service queue) and then uses this predicted state to look up the optimal power management action (e.g., voltage-frequency setting) from a precomputed policy table. The motivation for utilizing supervised learning in the form of a Bayesian classifier is to reduce the overhead of the PM which has to repetitively determine and assign voltage-frequency settings for each processor core in the system. Experimental results demonstrate that the proposed supervised learning based power management technique ensures system-wide energy savings under rapidly and widely varying workloads.Index Terms -Bayesian classification, dynamic power management, machine learning, multi-processor system, supervised learning

show abstract

Multicore thermal management with model predictive control

Zanini

Atienza

Benini

et al. 2009

2009 European Conference on Circuit Theory and Design

View full text Add to dashboard Cite

The goal of thermal management is to meet maximum operating temperature constraints, while at the same time tracking timevarying performance requirements. Current approaches avoid thermal violations by forcing abrupt operating points changes (e.g. processor shutdown), which cause sharp performance degradation. In this paper we aim at achieving a smooth thermal control action, that minimizes the variance of performance tracking error. We formulate this problem as a discrete-time optimal control problem, which can be solved using the theory and computational tools developed in the field of model-predictive control. Our optimization process considers the thermal profile of the system, its evolution over time, and time-varying workload requirements. Experimental results show that the proposed approach offers significant thermal balancing improvements over previous methods.

show abstract

Continuous Frequency Adjustment Technique Based on Dynamic Workload Prediction

Abstract: Abstract

Cited by 18 publications

References 14 publications

Workload Uncertainty Characterization and Adaptive Frequency Scaling for Energy Minimization of Embedded Systems

Workload Uncertainty Characterization and Adaptive Frequency Scaling for Energy Minimization of Embedded Systems

Supervised Learning Based Power Management for Multicore Processors

Multicore thermal management with model predictive control

Contact Info

Product

Resources

About