Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userPerformance-Aware Power Management in Embedded Controllers with Multiple-Voltage Processors
178
0
0.0
(
0
)
Added by
Feng Xia
Publication date
2008
fields
Informatics Engineering
and research's language is
English
Ask ChatGPT about the research
No Arabic abstract
The goal of this work is to minimize the energy dissipation of embedded controllers without jeopardizing the quality of control (QoC). Taking advantage of the dynamic voltage scaling (DVS) technology, this paper develops a performance-aware power management scheme for embedded controllers with processors that allow multiple voltage levels. The periods of control tasks are adapted online with respect to the current QoC, thus facilitating additional energy reduction over standard DVS. To avoid the waste of CPU resources as a result of the discrete voltage levels, a resource reclaiming mechanism is employed to maximize the CPU utilization and also to improve the QoC. Simulations are conducted to evaluate the performance of the proposed scheme. Compared with the optimal standard DVS scheme, the proposed scheme is shown to be able to save remarkably more energy while maintaining comparable QoC.
rate research
Read More
For microprocessors used in real-time embedded systems, minimizing power consumption is difficult due to the timing constraints. Dynamic voltage scaling (DVS) has been incorporated into modern microprocessors as a promising technique for exploring the trade-off between energy consumption and system performance. However, it remains a challenge to realize the potential of DVS in unpredictable environments where the system workload cannot be accurately known. Addressing system-level power-aware design for DVS-enabled embedded controllers, this paper establishes an analytical model for the DVS system that encompasses multiple real-time control tasks. From this model, a feedback control based approach to power management is developed to reduce dynamic power consumption while achieving good application performance. With this approach, the unpredictability and variability of task execution times can be attacked. Thanks to the use of feedback control theory, predictable performance of the DVS system is achieved, which is favorable to real-time applications. Extensive simulations are conducted to evaluate the performance of the proposed approach.
For better reliability and prolonged battery life, it is important that users and vendors understand the quality of charging and the performance of smartphone batteries. Considering the diverse set of devices and user behavior it is a challenge. In this work, we analyze a large collection of battery analytics dataset collected from 30K devices of 1.5K unique smartphone models. We analyze their battery properties and state of charge while charging, and reveal the characteristics of different components of their power management systems: charging mechanisms, state of charge estimation techniques, and their battery properties. We explore diverse charging behavior of devices and their users.
This paper presents, implements, and evaluates a power-regulation technique for multicore processors, based on an integral controller with adjustable gain. The gain is designed for wide stability margins, and computed in real time as part of the control law. The tracking performance of the control system is robust with respect to modeling uncertainties and computational errors in the loop. The main challenge of designing such a controller is that the power dissipation of program-workloads varies widely and often cannot be measured accurately; hence extant controllers are either ad hoc or based on a-priori modeling characterizations of the processor and workloads. Our approach is different. Leveraging the aforementioned robustness it uses a simple textbook modeling framework, and adjusts its parameters in real time by a system-identification module. In this it trades modeling precision for fast computations in the loop making it suitable for on-line implementation in commodity data-center processors. Consequently, the proposed controller is agnostic in the sense that it does not require any a-priori system characterizations. We present an implementation of the controller on Intels fourth-generation microarchitecture, Haswell, and test it on a number of industry benchmark programs which are used in scientific computing and datacenter applications. Results of these experiments are presented in detail exposing the practical challenges of implementing provably-convergent power regulation solutions in commodity multicore processors.
Dynamic voltage scaling (DVS) is one of the most effective techniques for reducing energy consumption in embedded and real-time systems. However, traditional DVS algorithms have inherent limitations on their capability in energy saving since they rarely take into account the actual application requirements and often exploit fixed timing constraints of real-time tasks. Taking advantage of application adaptation, an enhanced energy-aware feedback scheduling (EEAFS) scheme is proposed, which integrates feedback scheduling with DVS. To achieve further reduction in energy consumption over pure DVS while not jeopardizing the quality of control, the sampling period of each control loop is adapted to its actual control performance, thus exploring flexible timing constraints on control tasks. Extensive simulation results are given to demonstrate the effectiveness of EEAFS under different scenarios. Compared with the optimal pure DVS scheme, EEAFS saves much more energy while yielding comparable control performance.
In this paper, we compare the transient performance of a multi-terminal high-voltage DC (MTDC) grid equipped with a slack bus for voltage control to that of two distributed control schemes: a standard droop controller and a distributed averaging proportional-integral (DAPI) controller. We evaluate performance in terms of an H2 metric that quantifies expected deviations from nominal voltages, and show that the transient performance of a droop or DAPI controlled MTDC grid is always superior to that of an MTDC grid with a slack bus. In particular, by studying systems built up over lattice networks, we show that the H2 norm of a slack bus controlled system may scale unboundedly with network size, while the norm remains uniformly bounded with droop or DAPI control. We simulate the control strategies on radial MTDC networks to demonstrate that the transient performance for the slack bus controlled system deteriorates significantly as the network grows, which is not the case with the distributed control strategies.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
