TEEM: Online Thermal- and Energy-Efficiency Management on CPU-GPU MPSoCs

Isuwa, Samuel; Dey, Somdip; Singh, Amit Kumar; McDonald-Maier, Klaus

doi:10.23919/date.2019.8714794

Cited by 21 publications

(19 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To control the CPU temperature, different DVFS-based management methods have been proposed in HPC contexts [18]- [21]. Basireddy et al [18] present a workload-aware runtime energy management technique for efficient DVFS control.…”

Section: B Server Reliability Managementmentioning

confidence: 99%

“…However, these methods do not consider thermal and reliability management for the whole system. On the other hand, other studies [20], [21] consider performance, thermal and reliability in state-of-the-art heterogeneous multi-processor architectures. Nonetheless, these works do not consider the use of hybrid cache architectures and advanced control methods to set the DVFS level appropriately in very dynamic scenarios considering the different and variable reliability effects on resistive and main SRAM technologies.…”

Section: B Server Reliability Managementmentioning

confidence: 99%

See 1 more Smart Citation

COCKTAIL: Multicore Co-Optimization Framework With Proactive Reliability Management

Huang

Pahlevan

Zapater

et al. 2022

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

View full text Add to dashboard Cite

High performance computing (HPC) servers aim to meet an increase in the number and complexity of tasks and, consequently, to address the energy efficiency challenge. In addition to energy efficiency, it is essential to manage lifetime limitations of power-hungry components of servers (e.g., cores and cache), hence avoiding server failure before its lifetime period. Traditional approaches focus on either using hybrid caches to reduce the leakage power of traditional static randomaccess memory (SRAM) cache, and thus increase the energy efficiency, or the trade-off between the lifetime and performance of multi-core processors. However, these approaches fall short in terms of flexibility and applicability for HPC tasks in terms of multi-parametric optimization including quality-of-service (QoS), lifetime reliability, and energy efficiency. As a result, in this paper we propose COCKTAIL, a holistic strategy framework to jointly optimize the energy efficiency of multi-core server processors and tasks performance in the HPC context, while guaranteeing the lifetime reliability. First, we analyze the best cache technology among traditional SRAM and resistive random access memory (RRAM), within the context of hybrid cache architectures, to improve the energy efficiency and manage cache endurance limits with respect to tasks requirements. Second, we introduce a novel efficient proactive queue optimization policy to reorder HPC tasks for execution considering their end time and possible reliability effects on the use of the hybrid caches. Third, we present a dynamic model predictive control (MPC)-based reliability management method to maximize task performance, by controlling the frequency, temperature, and target lifetime of the server processor. Our results demonstrate that, while consuming similar energy, COCKTAIL provides up to 60% QoS improvement when compared to latest state-of-the-art energy optimization and reliability management techniques in the HPC context. Moreover, our strategy guarantees a design lifetime longer than 5 years for the whole HPC system.

show abstract

Section: B Server Reliability Managementmentioning

confidence: 99%

Section: B Server Reliability Managementmentioning

confidence: 99%

COCKTAIL: Multicore Co-Optimization Framework With Proactive Reliability Management

Huang

Pahlevan

Zapater

et al. 2022

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

View full text Add to dashboard Cite

show abstract

“…Isuwa et al [49] proposed a dynamic thermal-and energy-management approach for CPU-GPU based MP-SoCs by managing resources, frequency scaling and threadpartitioning of executing applications on CPU and GPU. The experiences were performed on Samsung Exynos 5422 MP-SoC.…”

Section: Reactive Approachesmentioning

confidence: 99%

Dynamic Energy and Thermal Management of Multi-core Mobile Platforms: A Survey

et al. 2020

Self Cite

View full text Add to dashboard Cite

Multi-core mobile platforms are on rise as they enable efficient parallel processing to meet ever-increasing performance requirements. However, since these platforms need to cater for increasingly dynamic workloads, efficient dynamic resource management is desired mainly to enhance the energy and thermal efficiency for better user experience with increased operational time and lifetime of mobile devices. This article provides a survey of dynamic energy and thermal management approaches for multi-core mobile platforms. These approaches do either proactive or reactive management. The upcoming trends and open challenges are also discussed.

show abstract

“…More advanced controllers make use of regression models to predict the future temperature and identify the operating point that achieves maximum performance, yet avoiding thermal violation. The model can be trained off-line on a set of representative benchmarks [27] or it can be continuously updated at run-time [28,29].…”

Section: Proactive Control Policiesmentioning

confidence: 99%

Performance Profiling of Embedded ConvNets under Thermal-Aware DVFS

2019

View full text Add to dashboard Cite

Convolutional Neural Networks (ConvNets) can be shrunk to fit embedded CPUs adopted on mobile end-nodes, like smartphones or drones. The deployment onto such devices encompasses several algorithmic level optimizations, e.g., topology restructuring, pruning, and quantization, that reduce the complexity of the network, ensuring less resource usage and hence higher speed. Several studies revealed remarkable performance, paving the way towards real-time inference on low power cores. However, continuous execution at maximum speed is quite unrealistic due to a fast increase of the on-chip temperature. Indeed, proper thermal management is paramount to guarantee silicon reliability and a safe user experience. Power management schemes, like voltage lowering and frequency scaling, are common knobs to control the thermal stability. Obviously, this implies a performance degradation, often not considered during the training and optimization stages. The objective of this work is to present the performance assessment of embedded ConvNets under thermal management. Our study covers the behavior of two control policies, namely reactive and proactive, implemented through the Dynamic Voltage-Frequency Scaling (DVFS) mechanism available on commercial embedded CPUs. As benchmarks, we used four state-of-the-art ConvNets for computer vision flashed into the ARM Cortex-A15 CPU. With the collected results, we aim to show the existing temperature-performance trade-off and give a more realistic analysis of the maximum performance achievable. Moreover, we empirically demonstrate the strict relationship between the on-chip thermal behavior and the hyper-parameters of the ConvNet, revealing optimization margins for a thermal-aware design of neural network layers.

show abstract

TEEM: Online Thermal- and Energy-Efficiency Management on CPU-GPU MPSoCs

Cited by 21 publications

References 25 publications

COCKTAIL: Multicore Co-Optimization Framework With Proactive Reliability Management

COCKTAIL: Multicore Co-Optimization Framework With Proactive Reliability Management

Dynamic Energy and Thermal Management of Multi-core Mobile Platforms: A Survey

Performance Profiling of Embedded ConvNets under Thermal-Aware DVFS

Contact Info

Product

Resources

About