Modular reinforcement learning for self-adaptive energy efficiency optimization in multicore system

Wang, Zhe; Tian, Zhongyuan; Xu, Jiang; Maeda, Rafael K. V.; Li, Haoran; Yang, Peng; Duong, Luan H. K.; Chen, Xuanqi

doi:10.1109/aspdac.2017.7858403

Cited by 28 publications

(19 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When there are only memory-intensive tasks in the workload, they scale down the chip frequency to save energy. In another trend, machine learning algorithms have been proposed to perform intelligent DVFS-based energy savings [2,14,15,17,23,32,33,35,36]. The authors of [36] used reinforcement learning, in which they took task characteristics and processor configurations to scale frequency for real-time systems.…”

Section: Related Workmentioning

confidence: 99%

Performance-Energy Trade-off in Modern CMPs

Abera

Balakrishnan

Kumar

2020

ACM Trans. Archit. Code Optim.

View full text Add to dashboard Cite

Chip multiprocessors (CMPs) are ubiquitous in all computing systems ranging from high-end servers to mobile devices. In these systems, energy consumption is a critical design constraint as it constitutes the most significant operating cost for computing clouds. Analogous to this, longer battery life continues to be an essential user concern in mobile devices. To optimize on power consumption, modern processors are designed with Dynamic Voltage and Frequency Scaling (DVFS) support at the individual core as well as the uncore level. This allows fine-grained control of performance and energy. For an n core processor with m core and uncore frequency choices, the total DVFS configuration space is now m (n+1) (with the uncore accounting for the + 1). In addition to that, in CMPs, the performance-energy trade-off due to core/uncore frequency scaling concerning a single application cannot be determined independently as cores share critical resources like the last level cache (LLC) and the memory. Thus, unlike the uni-processor environment, the energy consumption of an application running on a CMP depends not only on its characteristics but also on those of its co-runners (applications running on other cores). The key objective of our work is to select a suitable core and uncore frequency that minimizes power consumption while limiting application performance degradation within certain pre-defined limits (can be termed as QoS requirements). The key contribution of our work is a learning-based model that is able to capture the interference due to shared cache, bus bandwidth, and memory bandwidth between applications running on multiple cores and predict near-optimal frequencies for core and uncore.

show abstract

Section: Related Workmentioning

confidence: 99%

Performance-Energy Trade-off in Modern CMPs

Abera

Balakrishnan

Kumar

2020

ACM Trans. Archit. Code Optim.

View full text Add to dashboard Cite

show abstract

“…The approach addressed thermal cycling and average and peak temperatures simultaneously. An online DVFS control strategy based on core-level modular reinforcement learning to adaptively select appropriate operating frequencies for each individual core was proposed in [60]. An Q-learning based algorithm was proposed in [61] to identify V/F pairs for predicted workloads and given application performance requirements.…”

Section: Reinforcement Learning (Rl)mentioning

confidence: 99%

A Survey of Prediction and Classification Techniques in Multicore Processor Systems

Ababei

Moghaddam

2019

IEEE Trans. Parallel Distrib. Syst.

View full text Add to dashboard Cite

show abstract

“…To globally optimize the policy of voltage/frequency selection for improving energy efficiency in such scenarios, a core-level modular RL (MLR) based online DVFS, which explicitly considers the relationship between different cores, is proposed in [21]. This approach distributively applies modular Q-learning (MQL), one of the most commonly used MRL algorithm, to each core to learn the system behaviour.…”

Section: Dvfsmentioning

confidence: 99%

“…The machine learns to achieve a goal by trial-and-error interaction with a dynamic environment. RL algorithms are developed to find the optimal solution to sequential decision problem, and have been proven effective in a variety of problems from different areas [21]. RL is inspired by the trial-and-error method humans used for making decisions for millions of years.…”

Section: Introduction To Reinforcement Learningmentioning

confidence: 99%

“…The optimization of energy during application execution concerns to system level efforts. These efforts have tried to optimize energy by employing three essential ingredients: mapping [15][16][17], dynamic voltage and frequency scaling (DVFS) [18][19][20][21], and dynamic power management (DPM) [22][23][24][25][26]. The mapping defines assignment and ordering of the tasks and their communications onto resources of multi/many-core system in view of some optimization criteria such as compute performance and energy consumption.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Learning-Based Run-Time Power and Energy Management of Multi/Many-Core Systems: Current and Future Trends

Singh¹,

Leech²,

Basireddy³

et al. 2017

Journal of Low Power Electronics

View full text Add to dashboard Cite

Abstract-Multi/Many-core systems are prevalent in several application domains targeting different scales of computing such as embedded and cloud computing. These systems are able to fulfil the ever-increasing performance requirements by exploiting their parallel processing capabilities. However, effective power/energy management is required during system operations due to several reasons such as to increase the operational time of battery operated systems, reduce the energy cost of datacenters, and improve thermal efficiency and reliability. This article provides an extensive survey of learning-based run-time power/energy management approaches. The survey includes a taxonomy of the learning-based approaches. These approaches perform design-time and/or run-time power/energy management by employing some learning principles such as reinforcement learning. The survey also highlights the trends followed by the learning-based run-time power management approaches, their upcoming trends and open research challenges.

show abstract

Modular reinforcement learning for self-adaptive energy efficiency optimization in multicore system

Cited by 28 publications

References 16 publications

Performance-Energy Trade-off in Modern CMPs

Performance-Energy Trade-off in Modern CMPs

A Survey of Prediction and Classification Techniques in Multicore Processor Systems

Learning-Based Run-Time Power and Energy Management of Multi/Many-Core Systems: Current and Future Trends

Contact Info

Product

Resources

About