Cited By
View all- Senapati DSarkar AKarfa C(2021)HMDS: A Makespan Minimizing DAG Scheduler for Heterogeneous Distributed SystemsACM Transactions on Embedded Computing Systems10.1145/347703720:5s(1-26)Online publication date: 23-Sep-2021
Distributed power management of real-time applications on a GALS multiprocessor SOC
Abstract
It is generally desirable to reduce the power consumption of embedded systems. Dynamic Voltage and Frequency Scaling (DVFS) is a commonly applied technique to achieve power reduction at the cost of computational performance. Multiprocessor System on Chips (MPSoCs) can have multiple voltage and frequency domains, e.g. per-core. When DVFS is applied to real-time applications, the effects must be accounted for in the associated formal timing model.
In this work, we contribute our distributed multi-core run-time power-management technique for real-time dataflow applications that uses per-core lookup-tables to select low-power DVFS operating points that meet the application's timing requirement. We describe in detail how timing slack is observed locally at run-time on each core and is used to select a local DVFS operating point that meets the application's timing requirement. We further describe our static off-line formal analysis technique to generate these per-core lookup-tables that link timing slack to low-power DVFS operating points. We provide an experimental analysis of our proposed technique using an H.263 decoder application that is mapped onto an FPGA prototyped hardware platform.
References
[1]
S. Albers. Energy-efficient algorithms. Communications of the ACM, 53(5):86--96, 2010.
[2]
S. Albers et al. Speed scaling on parallel processors. In Proc. of the Nineteenth Annu. ACM Symp. on Parallel Algorithms and Architectures, pages 289--298, 2007.
[3]
A. Alimonda et al. A feedback-based approach to DVFS in data-flow applications. Computer-Aided Design of Integrated Circuits and Systems, IEEE Transactions on, 28(11):1691--1704, 2009.
[4]
L. Benini et al. A survey of design techniques for system-level dynamic power management. Very Large Scale Integration (VLSI) Systems, IEEE Transactions on, 8(3):299--316, 2000.
[5]
T. D. Burd and R. W. Brodersen. Energy efficient CMOS microprocessor design. In System Sciences, 1995. Proc. of the Twenty-Eighth Hawaii Int. Conf. on, pages 288--297, 1995.
[6]
M. Damavandpeyma et al. Throughput-constrained DVFS for scenario-aware dataflow graphs. In Real-Time and Embedded Technology and Applicat. Symp. (RTAS), 2013 IEEE 19th, pages 175--184, 2013.
[7]
P. de Langen. Energy Reduction Techniques for Caches and Multiprocessors. PhD thesis, Delft University of Technology, oct 2009.
[8]
P. de Langen and B. Juurlink. Leakage-aware multiprocessor scheduling. Journal of Signal Processing Systems, 57(1):73--88, 2009.
[9]
D. Dolev et al. Hex: scaling honeycombs is easier than scaling clock trees. In Proc. of the 25th ACM Symp. on Parallelism in algorithms and architectures, pages 164--175, 2013.
[10]
K. Goossens et al. Virtual execution platforms for mixed-time-criticality systems: The CompSOC architecture and design flow. ACM SIGBED Review, 10(3):23--34, 2013.
[11]
K. Govil et al. Comparing algorithm for dynamic speed-setting of a low-power CPU. In Proc. of the 1st Annu. Int. Conf. on Mobile computing and networking, pages 13--25, 1995.
[12]
M. Grant and S. Boyd. Cvx: Matlab software for dcp. http://cvxr.com/cvx, mar.
[13]
M. Grant and S. Boyd. Graph implementations for nonsmooth convex programs. In Recent Advances in Learning and Control, pages 95--110. Springer.
[14]
P. Huang et al. Energy efficient DVFS scheduling for mixed-criticality systems. In Embedded Software (EMSOFT), 2014 Int. Conf. on, pages 1--10, 2014.
[15]
S. Irani and K. R. Pruhs. Algorithmic problems in power management. ACM SIGACT News, 36(2):63--76, 2005.
[16]
P. Juang et al. Coordinated, distributed, formal energy management of chip multiprocessors. In Low Power Electronics and Design, 2005. ISLPED'05. Proc. of the 2005 Int. Symp. on, pages 127--130, 2005.
[17]
P. Kumar and L. Thiele. p-yds algorithm: An optimal extension of YDS algorithm to minimize expected energy for real-time jobs. In Embedded Software (EMSOFT), 2014 Int. Conf. on, pages 1--10, 2014.
[18]
T.-W. Lam et al. Energy efficient deadline scheduling in two processor systems. In Algorithms and Computation, pages 476--487. Springer, 2007.
[19]
T.-W. Lam et al. Competitive non-migratory scheduling for flow time and energy. In Proc. of the twentieth Annu. Symp. on Parallelism in algorithms and architectures, pages 256--264, 2008.
[20]
C. Lenzen et al. Clock synchronization with bounded global and local skew. In Foundations of Comput. Science, 2008. FOCS'08. IEEE 49th Annu. IEEE Symp. on, pages 509--518, 2008.
[21]
O. Moreira et al. Buffer sizing for rate-optimal single-rate data-flow scheduling revisited. IEEE Trans. Comput., pages 188--201, 2010.
[22]
A. Nelson et al. Power minimisation for real-time dataflow applications. In DSD, pages 117--124, 2011.
[23]
A. Nelson et al. Dataflow formalisation of real-time streaming applications on a composable and predictable multi-processor SOC. Journal of Systems Architecture, 2015.
[24]
M. Perner et al. Byzantine self-stabilizing clock distribution with HEX: Implementation, simulation, clock multiplication. In DEPEND 2013, The Sixth Int. Conf. on Dependability, pages 6--15, 2013.
[25]
K. Pruhs et al. Speed scaling of tasks with precedence constraints. Theory of Computing Systems, 43(1):67--80, 2008.
[26]
F. Yao et al. A scheduling model for reduced CPU energy. In Foundations of Comput. Science, pages 374--382, 1995.
Index Terms
- Distributed power management of real-time applications on a GALS multiprocessor SOC
Recommendations
Power Minimisation for Real-Time Dataflow Applications
DSD '11: Proceedings of the 2011 14th Euromicro Conference on Digital System DesignEnergy efficient execution of applications is important for many reasons, e.g. time between battery charges, device temperature. Voltage and Frequency Scaling (VFS) enables applications to be run at lower frequencies on hardware resources thereby ...
Communication-aware VFI partitioning for GALS-based networks-on-chip
The voltage/frequency island (VFI) design paradigm is a practical architecture for energy-efficient networks-on-chip (NoC) systems. In VFI-based NoC systems, each island can be operated with different voltage and clock frequency and thus it is important ...
Toward the optimal configuration of dynamic voltage scaling points in real-time applications
In real-time applications, compiler-directed dynamic voltage scaling (DVS) could reduce energy consumption efficiently, where compiler put voltage scaling points in the proper places, and the supply voltage and clock frequency were adjusted to the ...
Comments
Information & Contributors
Information
Published In
Sponsors
- SIGBED: ACM Special Interest Group on Embedded Systems
- SIGDA: ACM Special Interest Group on Design Automation
- IEEE: Institute of Electrical and Electronics Engineers
- SIGMICRO: ACM Special Interest Group on Microarchitectural Research and Processing
Publisher
IEEE Press
Publication History
Published: 04 October 2015
Check for updates
Author Tags
Qualifiers
- Research-article
Conference
Acceptance Rates
Overall Acceptance Rate 60 of 203 submissions, 30%
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- View Citations1Total Citations
- 69Total Downloads
- Downloads (Last 12 months)0
- Downloads (Last 6 weeks)0
Reflects downloads up to 01 Jan 2025
Other Metrics
Citations
Cited By
View all- Senapati DSarkar AKarfa C(2021)HMDS: A Makespan Minimizing DAG Scheduler for Heterogeneous Distributed SystemsACM Transactions on Embedded Computing Systems10.1145/347703720:5s(1-26)Online publication date: 23-Sep-2021
View Options
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Sign in