Dark silicon aware power management for manycore systems under dynamic workloads

Haghbayan²,

2015 IEEE/ACM International Symposium on Low Power Electronics and Design (ISLPED)

et al. 2015

Self Cite

Power management of NoC-based many-core systems with runtime application mapping becomes more challenging in the dark silicon era. It necessitates a multi-objective control approach to consider an upper limit on total power consumption, dynamic behaviour of workloads, processing elements utilization, per-core power consumption, and load on networkon-chip. In this paper, we propose a multi-objective dynamic power management method that simultaneously considers all of these parameters. Fine-grained voltage and frequency scaling, including near-threshold operation, and per-core power gating are utilized to optimize the performance. In addition, a disturbance rejecter is designed that proactively scales down activity in running applications when a new application commences execution, to prevent sharp power budget violations. Simulations of dynamic workloads and mixed time-critical application profiles show that our method is effective in honoring the power budget while considerably boosting the system throughput and reducing power budget violation, compared to the state-of-the-art power management policies.

show abstract

Section: Experimental Evaluationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Dynamic power management for many-core platforms in the dark silicon era: A multi-objective control approach

Haghbayan²,

2015 IEEE/ACM International Symposium on Low Power Electronics and Design (ISLPED)

et al. 2015

Self Cite

show abstract

“…Adaptive power capping techniques monitor the power consumption and actuate power knobs in a closed loop, in case of power consumption exceeding TDP. A PID controller based power management is presented in [13], where knob settings are actuated for power capping as per normalized gain of PID. Vega et.…”

Section: Introductionmentioning

confidence: 99%

“…The average service time for different knob combinations is shown in Figure 4(a). In case of using DVFS and PG knobs [13], per-application run-time increases forcing incoming applications to wait longer, resulting in high service time. The combination of DVFS and PG has relatively better service time using the power management algorithm, as in [5], [19].…”

Section: Introductionmentioning

confidence: 99%

Accuracy-Aware Power Management for Many-Core Systems Running Error-Resilient Applications

Haghbayan

et al. 2017

IEEE Trans. VLSI Syst.

Self Cite

Power capping techniques based on dynamic voltage and frequency scaling (DVFS) and power gating (PG) are oriented towards power actuation, compromising on performance and energy. Inherent error resilience of emerging application domains such as Internet-of-Things (IoT) and machine learning provide opportunities for energy and performance gains. Leveraging accuracy-performance trade-offs in such applications, we propose approximation (APPX) as another knob for close-looped power management, to complement power knobs with performance and energy gains. We design a power management framework, APPEND+, that can switch between accurate and approximate modes of execution subject to system throughput requirements. APPEND+ considers the sensitivity of the application to error to make disciplined alteration between levels of approximation such that performance is maximized while error is minimized. We implement a power management scheme that uses APPX, DVFS and PG knobs hierarchically. We evaluated our proposed approach over machine learning and signal processing applications along with two case studies on IoT -early warning score system and fall detection. APPEND+ yields 1.9× higher throughput, improved latency up to 5×, better performance per energy and dark silicon mitigation compared to state-of-the-art power management techniques over a set of applications ranging from high to no error resilience.

show abstract

A Perspective on Dark Silicon

The Dark Side of Silicon

Liljeberg

et al. 2017