All-digital embedded meters for on-line power estimation

Pagliari, Daniele Jahier; Peluso, Valentino; Calimera, Andrea; Macii, Enrico; Poncino, Massimo

doi:10.23919/date.2018.8342105

Cited by 10 publications

(14 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In particular, their implementation minimizes the performance overhead typically associated with the software-implemented schemes that, unfortunately, compete with the other applications for the CPU. Hardware solutions can also increase the timing resolution of the entire monitoring system to efficiently manage faster physical phenomena [5], [6], [7]. Note that our proposal targets the design of a controller that leverages on the all-digital scheme proposed in [6] as the online power monitor whose detailed description is out of the scope of this research.…”

Section: Background and Related Workmentioning

confidence: 99%

“…Nevertheless, it can be detrimental for the performance of embedded platforms, since part of the computing resources must be devoted to the computation of both the power estimate and the control action. Recent investigations [5], [6], [7] proposed all-digital power monitoring solutions that estimate the consumed power from the switching activity of selected signals at the microarchitectural level. However, there is a complete absence of a unified and general methodology to control different energy related aspects, also considering applications and Operating System requirements.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

All-Digital Control-Theoretic Scheme to Optimize Energy Budget and Allocation in Multi-Cores

Zoni

Cremona

Fornaciari

2020

IEEE Trans. Comput.

View full text Add to dashboard Cite

The Internet-of-Things (IoT) revolution fueled new challenges and opportunities to achieve computational efficiency goals. Embedded devices are required to execute multiple applications for which a suitable distribution of the computing power must be adapted at run-time. Such complex hardware platforms have to sustain the continuous acquisition and processing of data under severe energy budget constraints, since most of them are battery powered. The state-of-the-art offers several ad-hoc contributions to selectively optimize the performance considering aspects like energy, power, thermal or reliability. However, there is a need for a generic coordinated management strategy able to cope with all of these dimensions, while allowing the Operating System (OS) and the applications to "suggest" or constrain the actuation. This paper proposes a unified control-theoretic scheme to coordinate the design of energy-budget and energy allocation solutions for multi-cores. The proposed controller can work with any actuator and it can interact, at run-time, with both the applications and the OS to optimize the actuation signals steering the computing platform. Such control scheme offers the possibility to integrate any performance related policy in the form of an energy-allocation strategy, still ensuring the theoretic exponential stability of the overall controller if the actuation of the policy, coming from the OS and the applications, "is not too fast". To demonstrate the feasibility of our solution, we implemented the controller into a RISC multi-core running on the Xilinx Artix 100t FPGA device, available in the the Digilent Nexys4-DDR board. Results considering two actuators and both the quad-and the eight-core version of the considered computing platform, highlight the scalability of the proposed solution as well as an area overhead for the-all digital, on chip-controller limited to 0.86% (FFs) and 5.3% (LUTs) of the FPGA chip. We also considered a dynamic scenario validating the speed of the controller, where our framework has to face with modifications to the energy-allocation control policy carried out by the OS and the applications. The obtained results are collected by executing a huge mix of benchmarks and the statistical significance is accounted by executing each scenario 30 times. Such results are analyzed considering three quality metrics. First, the efficiency in exploiting the imposed budget (EF Fg) that is on average 98.27%. Second, the overflow of the actual average power consumption with respect to the assigned budget (OV Fg), which is limited to 1.43 mW on average. Last, the performance utility loss due to the control scheme that is limited to 1.87% on average.

show abstract

Section: Background and Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

All-Digital Control-Theoretic Scheme to Optimize Energy Budget and Allocation in Multi-Cores

Zoni

Cremona

Fornaciari

2020

IEEE Trans. Comput.

View full text Add to dashboard Cite

show abstract

“…It is clearly impossible to identify an accurate power model by means of the control signals only as suggested in [16]. On the other hand, we identified the power model according to the methodologies proposed in [11,15] that allow monitoring a portion of wider signals. In particular, the identified power model makes use of the two least significant bytes of the op0 operand to produce a power estimate.…”

Section: Motivational Examplementioning

confidence: 99%

“…Then, a representative set of the single-bit signals is selected to form the power model using the switching activity information extracted from the simulation of a set of benchmarks. A similar approach is discussed in [15], targeting single bit flip-flops instead of the corresponding driven signals.…”

Section: Related Workmentioning

confidence: 99%

“…In contrast, the most precise switching information correlating with the power consumption can be extracted by monitoring the toggle activity of each signal in the target platform at Register Transfer Level (RTL) or the corresponding activity of the driving logic cells. While the high number of signals to be monitored makes such solution infeasible, the possibility to remove the drawbacks of Power Monitoring Counter schemes motivates the proposal of a number of all-digital power monitoring solutions which try to optimally balance the number of monitored RTL signals and the accuracy of the power estimate [11,14,15].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation