Vishnu Swaminathan
Krishnendu Chakrabarty
Abstract
Software-controlled (or dynamic) power management (DPM) in embedded systems has emerged as an attractive alternative to inflexible hardware solutions. However, DPM via I/O device scheduling for hard real-time systems has received relatively little attention. In this paper, we present an offline I/O device scheduling algorithm called energy-optimal device scheduler (EDS). For a given set of jobs, it determines the start time of each job such that the energy consumption of the I/O devices is minimized. EDS also ensures that no real-time constraint is violated. The device schedules are provably energy optimal under hard real-time job deadlines. Temporal and energy-based pruning are used to reduce the search space significantly. Since the I/O device scheduling problem is NP-complete, we also describe a heuristic called maximum device overlap (MDO) to generate near-optimal solutions in polynomial time. We present experimental results to show that EDS and MDO reduce the energy consumption of I/O devices significantly for hard real-time systems.
- AbouGhazaleh, N., Mosse, D., Childers, B., and Melham, R. 2001. Toward the placement of power management points in real-time applications. In Proceedings Workshop on Compilers and Operating Systems for Low Power (COLP). Google Scholar
- Advanced Configuration and Power Interface (ACPI). http://www.teleport.com/~acpi.Google Scholar
- Alidina, M., Monteiro, J., Devadas, S., Ghosh, A., and Papaefthymiou, M. 1994. Precomputation-based sequential logic optimization for low power. IEEE Trans. VLSI Syst. 2, 426--436. Google Scholar
- AMD Am79C874 NetPHY-1LP Low-Power 10/100 Tx/Rx Ethernet Transceiver Technical Datasheet.Google Scholar
- Baker, T. P. 1991. Stack-based scheduling of real-time processes. Real-Time Syst. J. 3, 1, 67--100. Google Scholar
- Benini, L., Bogliolo, A., Paleologo, G. A., and De Micheli, G. 1999. Policy optimization for dynamic power management. IEEE Trans. Comput.-Aided Des. 16, 813--833. Google Scholar
- Buttazzo, G. C. 1997. Hard Real-time Computing Systems: Predictable Scheduling Algorithms and Applications, Kluwer Academic Publishers, Norwell, MA. Google Scholar
- Chung, E.-Y., Benini, L., and De Micheli, G. 1999. Dynamic power management using adaptive learning tree. In Proceedings of the International Conference on Computer-Aided Design. 274--279. Google Scholar
- Advanced Configuration and Power Interface (ACPI). http://www.teleport.com/~acpi.Google Scholar
- Garey, M. R. and Johnson, D. S. 1977. Two-processor scheduling with start times and deadlines. SIAM J. Comput. 6, 416--426.Google Scholar
- Golding, R., Bosh, P., Staelin, C., Sullivan, T., and Wilkes, J. 1995. Idleness is not sloth. In Proceedings of the Usenix Technical Conference on UNIX and Advanced Computing Systems. 201--212. Google Scholar
- Hong, I., Potkonjak, M., and Srivastava, M. B. 1998. On-line scheduling of hard real-time tasks on variable-voltage processor. In Proceedings of the International Conference on Computer-Aided Design. 653--656. Google Scholar
- Hwang, C. and C.-H.Wu, A. 1997. A predictive system shutdown method for energy saving of event-driven computation. In Proceedings of the International Conference on Computer-Aided Design. 28--32. Google Scholar
- Irani, S., Shukla, S. and Gupta, R. 2002. Competitive analysis of dynamic power management strategies for systems with multiple power states. In Proceedings of the Design Automation and Test in Europe (DATE) Conference. 117--123. Google Scholar
- Ishihara, T. and Yasuura, H. 1998. Voltage scheduling problem for dynamically variable voltage processors. In Proceedings of the International Symposium on Low-Power Electronics and Design. 197--202. Google Scholar
- Katcher, D., Arakawa, H., and Strosnider, J. 1993. Engineering and analysis of fixed priority schedulers. IEEE Trans. Software Eng. 19, 920--934. Google Scholar
- Kim, W., Kim, J., and Min, S. L. 2002. A dynamic voltage scaling algorithm for dynamic priority hard real-time systems using slack time analysis. In Proceedings of the Design Automation and Test Conference in Europe. 788--795. Google Scholar
- Lawler, E. L. 1973. Optimal sequencing of a single machine subject to precedence constraints. J. Management Sci. 19, 544--546.Google Scholar
- Li, D., Chou, P., and Bagerzadeh, N. 2002. Mode selection and mode-dependency modeling for power-aware embedded systems. In Proceedings of the Asia South Pacific Design Automation Conference. 697--704. Google Scholar
- Liu, C. L., and Layland, J. 1973. Scheduling algorithms for multiprogramming in a hard real-time environment. J. ACM. 20, 1, 46--61. Google Scholar
- Liu, J. W. S. 2000. Real-time Systems, Prentice-Hall, Upper Saddle River, NJ. Google Scholar
- Liu, J., Chou, P. H., and Bagerzadeh, N. 2002. Communication speed selection for embedded systems with networked voltage-scalable processors. In Proceedings of the International Symposium on Hardware/Software Codesign (CODES). 169--174. Google Scholar
- Lu, Y.-H., Benini, L., and De Micheli, G. 2000. Low-power task scheduling for multiple devices. In Proceedings of the International Workshop on Hardware/Software Codesign. 39--43. Google Scholar
- Lu, Y.-H., Benini, L., and De Micheli, G. 2000. Operating system directed power reduction. In Proceedings of the International Conference on Low-Power Electronics and Design. 37--42. Google Scholar
- Luo, J. and Jha, N. K. 2001. Battery-aware static scheduling for distributed real-time embedded systems. In Proceedings of the Design Automation Conference. 444--449. Google Scholar
- Luo, J. and Jha, N. 2002. Static and dynamic variable voltage scheduling algorithms for real-time heterogeneous distributed embedded systems. In Proceedings of the International Conference on VLSI Design. 719--726. Google Scholar
- Quan, G. and Hu, X. 2001. Energy efficient fixed-priority scheduling for real-time systems on variable voltage processors. In Proceedings of the Design Automation Conference. 828--833. Google Scholar
- Quan, G. and Hu, X. 2002. Minimum-energy fixed-priority scheduling for variable-voltage processor. In Proceedings of the Design Automation and Test in Europe (DATE) Conference. 782--788. Google Scholar
- Shin, Y. and Choi, K. 1999. Power conscious fixed priority scheduling for hard real-time systems. In Proceedings of the Design Automation Conference. 134--139. Google Scholar
- Sha, L., Rajkumar, R., and Lehoczky, J. P. 1990. Priority inheritance protocols: An approach to real-time synchronization. IEEE Trans. Comput. 39, 1175--1185. Google Scholar
- Shin, Y., Choi, K., and Sakurai, T. 2000. Power optimization of real-time embedded systems on variable speed processors. In Proceedings of the International Conference on Computer-Aided Design. 365--368. Google Scholar
- Simunic, T., Benini, L., Glynn, P., and De Micheli, G. 2001. Event driven power management. IEEE Trans. Comput.-Aided Des. 20, 840--857. Google Scholar
- Swaminathan, V. and Chakrabarty, K. 2003. Energy-conscious, deterministic I/O device scheduling in hard real-time systems. IEEE Transaction on Computer-Aided Design Integrated Circuits and Systems 22 (July), 847--858. Google Scholar
- TMS320C6411 Power Consumption Summary. Available on-line at www-s.ti.com/sc/techlit/spra373.Google Scholar
- Xu, J. and Parnas, D. L. 2000. Priority scheduling vs. pre-run-time scheduling. Int. J. Time-Critical Comput. Syst. 18, 7--23. Google Scholar
- Yao, F., Demers, A., and Shenker, S. 1995. A scheduling model for reduced CPU energy. In Proceedings of the IEEE Annual Foundations of Computer Science. 374--382. Google Scholar
- Zhang, Y., Hu, X., and Chen, D. 2002. Task scheduling and voltage selection for energy minimization. In Proceedings of the Design Automation Conference. 183--188. Google Scholar
Index Terms
- Pruning-based, energy-optimal, deterministic I/O device scheduling for hard real-time systems
Recommendations
Pruning-based energy-optimal device scheduling for hard real-time systems
CODES '02: Proceedings of the tenth international symposium on Hardware/software codesignDynamic Power Management (DPM) provides a simple, elegant and flexible method for reducing energy consumption in embedded real-time systems. However, I/O-centric DPM techniques have been studied largely for non-real-time environments. We present an ...
A Pre-Run-Time Scheduling Algorithm for Hard Real-Time Systems
Process scheduling, an important issue in the design and maintenance of hard real-time systems, is discussed. A pre-run-time scheduling algorithm that addresses the problem of process sequencing is presented. The algorithm is designed for multiprocessor ...
Energy-conscious, deterministic I/O device scheduling in hard real-time systems
Energy consumption is an important design parameter for embedded and portable systems. Software-controlled (or dynamic) power management (DPM) has emerged as an attractive alternative to inflexible hardware solutions. However, DPM via I/O device ...
Reviews
Vassilios A. Chouliaras
Swaminathan and Chakrabarty address the issue of optimal job scheduling in a hard real-time computer system, via the use of a novel, off-line dynamic power management (DPM) scheme for computer system input/output (I/O). The paper elaborates briefly on the multiple application domains in which embedded systems are utilized, and the well-known benefits of reducing power consumption. Subsequently, it discusses static (leakage) and dynamic (switching) power in the context of integrated circuits. The authors then introduce the concept of dynamic power management, which allows a system to make runtime decisions to take advantage of periods of inactivity and computational demands. The authors very correctly point out that DPM, at the operating system (OS) level, has been used extensively since 1997 via the advanced configuration and power interface (ACPI). Furthermore, they discuss present-day computer systems, and identify (quite correctly) both the central processing unit (CPU) and the I/O on desktop computer systems as being extremely important systems for power optimization. The introductory section concludes with a brief mention of the contributed, nondeterministic polynomial time (NP) complete algorithm, energy-optimal device scheduler (EDS), and the implementation of a polynomial-time maximum device overlap (MDO) heuristic to significantly reduce I/O device power consumption. The "Prior Work" section in the paper sets the scene for the contributed EDS algorithm, and I/O DPM is classified as timeout-based, predictive, and stochastic. Mention is made of the very important (and reasonably recent) development in the embedded microprocessor industry, dynamic voltage scaling (DVS). The authors then introduce notation and formalize the problem, and demonstrate that it is NP-complete. They introduce their methodology, based on energy pruning, through a very informative example section that takes the reader through the EDS algorithm; I found this section to be particularly useful and informative in my understanding of the work. EDS tries to find an optimal solution for an NP-complete problem, which potentially leads to very long runtimes and excessive memory requirements. The authors therefore introduce a heuristic method to generate near-optimal solutions in polynomial time, which they call MDO. Unfortunately, MDO does not address the issue of blocking, but the authors do mention that resource allocation by an algorithm such as the stack resource policy (SRP) will prevent this from arising. In any case, MDO is a significant step in computing a provably NP-complete problem in polynomial time. The experimentation section discusses two experiments, with the first consisting of three devices: a network interface card, a digital signal processor (DSP), and a hard disk drive (HDD) with multiple sleep states. The EDS algorithm in this case arrives at an optimal 33 percent energy reduction (presumably with respect to keeping all three devices powered up all the time). The heuristic-based MDO arrives at a solution that is only four percent worse than the optimal (EDS) solution, clearly demonstrating the strength of the heuristic method. Overall, I found the work to be very satisfactory and informative. There are very interesting synergies between I/O DPM and CPU DVS, and I hope the authors will pursue a study on that front in a forthcoming contribution. Online Computing Reviews Service
Access critical reviews of Computing literature here
Become a reviewer for Computing Reviews.
Comments