Computer Science Department University of Pittsburgh 1 Evaluating a DVS Scheme for Real-Time Embedded Systems Ruibin Xu, Daniel Mossé and Rami Melhem.

Slides:



Advertisements
Similar presentations
Marcus T. Schmitz and Bashir M. Al-Hashimi
Advertisements

Feedback EDF Scheduling Exploiting Dynamic Voltage Scaling Yifan Zhu and Frank Mueller Department of Computer Science Center for Embedded Systems Research.
Power Aware Scheduling for AND/OR Graphs in Multi-Processor Real-Time Systems Dakai Zhu, Nevine AbouGhazaleh, Daniel Mossé and Rami Melhem PARTS Group.
Pinwheel Scheduling for Power-Aware Real-Time Systems Gaurav Chitroda Komal Kasat Nalini Kumar.
Energy-efficient Task Scheduling in Heterogeneous Environment 2013/10/25.
Zhou Peng, Zuo Decheng, Zhou Haiying Harbin Institute of Technology 1.
1 EE5900 Advanced Embedded System For Smart Infrastructure Energy Efficient Scheduling.
1 “Scheduling with Dynamic Voltage/Speed Adjustment Using Slack Reclamation In Multi-processor Real-Time Systems” Dakai Zhu, Rami Melhem, and Bruce Childers.
Real- time Dynamic Voltage Scaling for Low- Power Embedded Operating Systems Written by P. Pillai and K.G. Shin Presented by Gaurav Saxena CSE 666 – Real.
Introduction and Background  Power: A Critical Dimension for Embedded Systems  Dynamic power dominates; static /leakage power increases faster  Common.
Power Aware Real-time Systems Rami Melhem A joint project with Daniel Mosse, Bruce Childers, Mootaz Elnozahy.
Minimizing Expected Energy Consumption in Real-Time Systems through Dynamic Voltage Scaling Ruibin Xu, Daniel Mosse’, and Rami Melhem.
1 Stochastic Event Capture Using Mobile Sensors Subject to a Quality Metric Nabhendra Bisnik, Alhussein A. Abouzeid, and Volkan Isler Rensselaer Polytechnic.
Aleksandra Tešanović Low Power/Energy Scheduling for Real-Time Systems Aleksandra Tešanović Real-Time Systems Laboratory Department of Computer and Information.
Investigating the Effect of Voltage- Switching on Low-Energy Task Scheduling in Hard Real-Time Systems Paper review Presented by Chung-Fu Kao.
System-Wide Energy Minimization for Real-Time Tasks: Lower Bound and Approximation Xiliang Zhong and Cheng-Zhong Xu Dept. of Electrical & Computer Engg.
Linear Solution to Scale and Rotation Invariant Object Matching Professor: 王聖智 教授 Student : 周 節.
Processor Frequency Setting for Energy Minimization of Streaming Multimedia Application by A. Acquaviva, L. Benini, and B. Riccò, in Proc. 9th Internation.
1 Assessment of Imprecise Reliability Using Efficient Probabilistic Reanalysis Farizal Efstratios Nikolaidis SAE 2007 World Congress.
Soner Yaldiz, Alper Demir, Serdar Tasiran Koç University, Istanbul, Turkey Paolo Ienne, Yusuf Leblebici Swiss Federal Institute of Technology (EPFL), Lausanne,
Introduction Due to the recent advances in smart grid as well as the increasing dissemination of smart meters, the electricity usage of every moment in.
VOLTAGE SCHEDULING HEURISTIC for REAL-TIME TASK GRAPHS D. Roychowdhury, I. Koren, C. M. Krishna University of Massachusetts, Amherst Y.-H. Lee Arizona.
Abhilash Thekkilakattil, Radu Dobrin, Sasikumar Punnekkat Mälardalen Real-time Research Center, Mälardalen University Västerås, Sweden Preemption Control.
Baoxian Zhao Hakan Aydin Dakai Zhu Computer Science Department Computer Science Department George Mason University University of Texas at San Antonio DAC.
Real-time Object Image Tracking Based on Block- Matching Algorithm ECE 734 Hsiang-Kuo Tang Tai-Hsuan Wu Ying-Tien Lin.
Energy-Efficient Soft Real-Time CPU Scheduling for Mobile Multimedia Systems Wanghong Yuan, Klara Nahrstedt Department of Computer Science University of.
Low Power Design for Real-Time Systems Low power (energy) consumption is a key design for embedded systems Battery’s life during operation Reliability.
1 EE5900 Advanced Embedded System For Smart Infrastructure Energy Efficient Scheduling.
Dynamic Slack Reclamation with Procrastination Scheduling in Real- Time Embedded Systems Paper by Ravindra R. Jejurikar and Rajesh Gupta Presentation by.
Probabilistic Preemption Control using Frequency Scaling for Sporadic Real-time Tasks Abhilash Thekkilakattil, Radu Dobrin and Sasikumar Punnekkat.
Scheduling policies for real- time embedded systems.
1 Customer-Aware Task Allocation and Scheduling for Multi-Mode MPSoCs Lin Huang, Rong Ye and Qiang Xu CHhk REliable computing laboratory (CURE) The Chinese.
1 Distributed Energy-Efficient Scheduling for Data-Intensive Applications with Deadline Constraints on Data Grids Cong Liu and Xiao Qin Auburn University.
An Energy-Efficient Hypervisor Scheduler for Asymmetric Multi- core 1 Ching-Chi Lin Institute of Information Science, Academia Sinica Department of Computer.
Euro-Par, A Resource Allocation Approach for Supporting Time-Critical Applications in Grid Environments Qian Zhu and Gagan Agrawal Department of.
The Fast Optimal Voltage Partitioning Algorithm For Peak Power Density Minimization Jia Wang, Shiyan Hu Department of Electrical and Computer Engineering.
Abhilash Thekkilakattil, Radu Dobrin, Sasikumar Punnekkat Mälardalen Real-time Research Center, Mälardalen University Västerås, Sweden Towards Preemption.
Multi-Criteria Routing in Pervasive Environment with Sensors Santhanakrishnan, G., Li, Q., Beaver, J., Chrysanthis, P.K., Amer, A. and Labrinidis, A Department.
Company name KUAS HPDS A Realistic Variable Voltage Scheduling Model for Real-Time Applications ICCAD Proceedings of the 2002 IEEE/ACM international conference.
Hard Real-Time Scheduling for Low- Energy Using Stochastic Data and DVS Processors Flavius Gruian Department of Computer Science, Lund University Box 118.
Efficient Energy Management Protocol for Target Tracking Sensor Networks X. Du, F. Lin Department of Computer Science North Dakota State University Fargo,
Dynamic Voltage Frequency Scaling for Multi-tasking Systems Using Online Learning Gaurav DhimanTajana Simunic Rosing Department of Computer Science and.
An Energy-efficient Task Scheduler for Multi-core Platforms with per-core DVFS Based on Task Characteristics Ching-Chi Lin Institute of Information Science,
DTM and Reliability High temperature greatly degrades reliability
June 30 - July 2, 2009AIMS 2009 Towards Energy Efficient Change Management in A Cloud Computing Environment: A Pro-Active Approach H. AbdelSalamK. Maly.
OPERATING SYSTEMS CS 3530 Summer 2014 Systems and Models Chapter 03.
Rounding scheme if r * j  1 then r j := 1  When the number of processors assigned in the continuous solution is between 0 and 1 for each task, the speed.
Shibo He 、 Jiming Chen 、 Xu Li 、, Xuemin (Sherman) Shen and Youxian Sun State Key Laboratory of Industrial Control Technology, Zhejiang University, China.
Multimedia Computing and Networking Jan Reduced Energy Decoding of MPEG Streams Malena Mesarina, HP Labs/UCLA CS Dept Yoshio Turner, HP Labs.
Computer Science Department University of Pittsburgh Multiple-Resource Periodic Scheduling Problem: how much fairness is necessary? Dakai Zhu, Daniel Mossé.
CprE 458/558: Real-Time Systems (G. Manimaran)1 Energy Aware Real Time Systems - Scheduling algorithms Acknowledgement: G. Sudha Anil Kumar Real Time Computing.
VLSI Design & Embedded Systems Conference January 2015 Bengaluru, India Few Good Frequencies for Power-Constrained Test Sindhu Gunasekar and Vishwani D.
Workload Clustering for Increasing Energy Savings on Embedded MPSoCs S. H. K. Narayanan, O. Ozturk, M. Kandemir, M. Karakoy.
Linear Solution to Scale and Rotation Invariant Object Matching Hao Jiang and Stella X. Yu Computer Science Department Boston College.
Toward Reliable and Efficient Reporting in Wireless Sensor Networks Authors: Fatma Bouabdallah Nizar Bouabdallah Raouf Boutaba.
A stochastic scheduling algorithm for precedence constrained tasks on Grid Future Generation Computer Systems (2011) Xiaoyong Tang, Kenli Li, Guiping Liao,
Determining Optimal Processor Speeds for Periodic Real-Time Tasks with Different Power Characteristics H. Aydın, R. Melhem, D. Mossé, P.M. Alvarez University.
CSE 340 Computer Architecture Summer 2016 Understanding Performance.
Improving Dynamic Voltage Scaling Algorithms with PACE Jacob R. LorchAlan Jay Smith University of California Berkeley June 18, 2001 To make the most of.
Optimization of Time-Partitions for Mixed-Criticality Real-Time Distributed Embedded Systems Domițian Tămaș-Selicean and Paul Pop Technical University.
OPERATING SYSTEMS CS 3502 Fall 2017
Jacob R. Lorch Microsoft Research
Babak Sorkhpour, Prof. Roman Obermaisser, Ayman Murshed
Department of Electrical & Computer Engineering
Energy Efficient Scheduling in IoT Networks
Flavius Gruian and Krzysztof Kuchcinski
Jian-Jia Chen and Tei-Wei Kuo
FAST: Frequency-Aware Static Timing Analysis
Presentation transcript:

Computer Science Department University of Pittsburgh 1 Evaluating a DVS Scheme for Real-Time Embedded Systems Ruibin Xu, Daniel Mossé and Rami Melhem

Computer Science Department University of Pittsburgh 2 Introduction Energy conservation is important for real- time embedded systems Dynamic Voltage Scaling (DVS) is effective in power management A popular problem: minimizing energy consumption while meeting the deadlines

Computer Science Department University of Pittsburgh 3 Focus Frame-based systems that execute variable workloads The problem becomes minimizing the expected energy consumption while meeting the deadlines …… time Frame length

Computer Science Department University of Pittsburgh 4 A New DVS Scheme (MEEC) simplified problem original problem optimal solution practical solution relax fix Evaluations efficient algorithm emsoft’05 parc’05

Computer Science Department University of Pittsburgh 5 Task and System Model N periodic tasksT 1, T 2, …, T N to be executed consecutively in each frame The power function is p(f) = c 0 +c 1 f α

Computer Science Department University of Pittsburgh 6 Review of Existing Schemes slack Proportional Scheme Greedy Scheme Statistical Scheme

Computer Science Department University of Pittsburgh 7 The MEEC Scheme Incorporates the variability of the tasks into the speed schedule The variability of the tasks are captured by the probability density function of the workload of the tasks Aims to minimize the expected energy consumption in the system workload probability

Computer Science Department University of Pittsburgh 8 The MEEC Scheme slack β1dβ1d (1-β 1 )d d β1β1 β2β2 β3β3 β4β4

Computer Science Department University of Pittsburgh 9 An Important Property The optimal expected energy consumption for dd … are Both are proportional to 1/d 2

Computer Science Department University of Pittsburgh 10 Computing β i β 4 =100% β 3 =xx% vs. T1T1 T2T2 T3T3 T4T4 β 2 =xx% vs. β 1 =xx%

Computer Science Department University of Pittsburgh 11 Applying PACE PACE is a technique in which the execution speed is gradually increased as the task progresses

Computer Science Department University of Pittsburgh 12 The MEEC Scheme The β values (optimal) are computed based on the assumption of unrestricted continuous frequency We need to deal with:  Minimum and maximum speed restriction  Discrete speed We have solutions and will use simulation to test them

Computer Science Department University of Pittsburgh 13 Evaluations – Power models Synthetic processor  Strictly conforms to p(f)=f 3  10 frequencies: 100MHz, 200MHz,…, 1000MHz Intel Xscale  Power numbers from Intel datasheets  p(f) = (f/1000) 3

Computer Science Department University of Pittsburgh 14 Evaluation – Synthetic Workload We simulated systems that have 5,10,15,20 tasks The WCEC of each task is randomly generated from 10M to 1G cycles The probability distribution of each task is randomly chosen from 6 representative distributions Frame length

Computer Science Department University of Pittsburgh 15 Evaluation – Synthetic Workload We evaluated 8 schemes  Proportional with and without PACE  Greedy with and without PACE  Statistical with and without PACE  MEEC with and without PACE We simulated 100,000 frames and computed the average energy consumption per frame for each scheme

Computer Science Department University of Pittsburgh 16 Results – Synthetic Workload For synthetic CPU, the best scheme is always MEEC (with or without PACE), but MEEC with PACE is only better than MEEE without PACE 13.6% of the time with an average saving of 1.2% For Intel Xscale, the best scheme is always MEEC without PACE Conclusion: PACE is not recommended in the MEEC scheme

Computer Science Department University of Pittsburgh 17 Why PACE Is Not Good in MEEC scheme? PACE (under the assumption of unrestricted continuous frequency) PACE (discrete frequency) fix β values compute Can differ a lot

Computer Science Department University of Pittsburgh 18 Results – Synthetic Workload

Computer Science Department University of Pittsburgh 19 Evaluation – Automatic Target Recognition (ATR) The ATR application does pattern matching of targets in images The regions of interest (ROI) in the image are detected and each ROI is compared with all the templates Image processing time is proportional to the number of ROIs

Computer Science Department University of Pittsburgh 20 Evaluation – Automatic Target Recognition (ATR) A front-end is responsible for collecting images and send them to the back-end periodically for target recognition This application can be modeled as a frame-based real-time system in which all the tasks have the same workload distribution front-end back-end ……

Computer Science Department University of Pittsburgh 21 Evaluation – Automatic Target Recognition (ATR) Simulation setup  Use Intel Xscale  The period is 100ms  The front-end sends 1 to 6 images to the back-end  The number of ROIs in an image varies from 1 to 8  The back-end precomputes 6 speed schedules

Computer Science Department University of Pittsburgh 22 Results - Automatic Target Recognition (ATR)

Computer Science Department University of Pittsburgh 23 Summary In this paper, we demonstrate and evaluate a new DVS scheme that aims to minimize the expected energy consumption in the system

Computer Science Department University of Pittsburgh 24 Conclusions The MEEC scheme achieves significant energy savings over the existing schemes Using only static information or aggregating dynamic information, even with probabilistic techniques, will not produce as good results as when dynamic information for each task in considered separately

Computer Science Department University of Pittsburgh 25 Thank you

Computer Science Department University of Pittsburgh 26 A Simple Example 3 tasks, the frame length is 14 time units For the CPU, c 0 =0, c 1 =1, f min =0, and f max =1