在本研究中專注於一個一直被忽略的問題，而Power Transition Overhead就是一個典 型的例子。
在本研究中使用SA1110的電源參數，探討在轉換的過程中所造成時間的延遲和能量的耗 損，透過數學模型與電子電路的物理性質已證明了這種現像的存在，在文中提出了一個 演算法主要包含三個部份:Slack Estimation， Return of Investment與Power Control，
能夠有效的降低許多不必要的轉換，在大多數的場合裡以整體的耗電量而言也有節省的 趨 勢 。 此 外 ， 對 即 時 性 也 有 較 佳 的 保 障 。 針 對 減 少 的 耗 損 而 言 ， 以 一 般 的 Statically-Schedule EDF的演算法做比較有30%的減少，與較佳且先進的Look-Ahead EDF演算法做比較，使用本文的前後也有2%到5%的降低。 雖然在CPU接近滿載(超過90%) 時的耗電量會稍微高一點，但在即時系統中，應該還是符合絕大多數的工作範圍。
因此，這項議題將來深入研究將有一個主要的方面，就是slack estimation，如果能夠 增進slack預測的可靠度，那麼將可以加以利用，使得能夠更貼近系統實際的能量需要 以求取最佳的利用率，降低耗電量。
參考文獻
[1] A. Vahdat， A. R. Lebeck and C. S. Ellis. “Every Joule is Precious: A Case for Revisiting Operating System Design for Energy Efficiency.” In the 9th ACM SIGOPS European Workshop， September 2000.
[2] P. J. M. Havinga and G. J. M. Smith. “Design Techniques for Low-power Systems.” Journal of Systems Architecture. Vol. 46:1， 2000
[3] M. Pedram. “Power Minimization in IC Design: Principles and Applications.”
ACM Transactions on Design Automation of Electronics Systems. 1:1 - pp.
3-56， January 1996.
[4] Intel， Microsoft， and Toshiba. “Advanced Configuration and Power Management Interface (ACPI) Specification”， 1999.
www.intel.com/ial/WfM/design/pmdt/acpidesc.htm.
[5] M. Srivastava， A. P. Chandrakasan and R. W. Brodersen. “Predictive System Shutdown and other Architectural Techniques for Energy Efficient
Programmable Computation.” IEEE Trans. on VLSI Systems， 4(1): 42-55，
1996.
[6] Intel， Mobile Intel Pentium III Processor-M
http://www.intel.com/products/notebook/processors/pentiumiii-m/index.ht m?iid=sr+iii&
[7] M. Weiser， B. Welch， A. Demers， and S. Shenker. “Scheduling for Reduced CPU Energy.” In Proceedings of USENIX Symposium on Operating Systems Design and Implementation， 1994.
[8] E. Chan， K. Govil， and H. Wasserman. “Comparing Algorithms for Dynamic Speed-setting of a Low-Power CPU.” In Proceedings of the First ACM International Conference on Mobile Computing and Networking (MOBICOM 95)，
November 1995.
[9] J. Pouwelse， K. Langendoen and H. Sips. Dynamic Voltage Scaling on a Low-Power Microprocessor. 7th International Conference on Mobile Computing and Networking (MOBICOM)， Rome， Italy， July 2001.
[10] D. Shin， J. Kim and S. Lee， "Intra-Task Voltage Scheduling for Low-Energy Hard Real-Time Applications，" IEEE Design and Test of Computers， March 2001
[11]N. AbouGhazaleh， D. Mossé， B. Childers and R. Melhem “Toward The Placement of Power Management Points in Real Time Applications”， COLP'01 (Workshop on Compilers and Operating Systems for Low Power)， Barcelona， Spain， 2001 [12]L.H. Chandrasena and M.J. Liebelt. “A Rate Selection Algorithm for
Quantized Undithered Dynamic Supply Voltage Scaling，” ISLPED’00:
International Symposium on Low Power Electronics and Design， pp. 213-215，
July 2000.
[13]J.Chang and M. Pedram. “Energy Minimization Using Multiple Supply Voltages，” IEEE Transactions on Very Large Scale Integration Systems，
Vol.5， No.4， pp. 436-443， December 1997.
[14]I. Hong， et al. “Power Optimization of Variable-Voltage Core-Based Systems，” IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems， Vol.18， No.12， pp. 1702-1714， December 1999.
[15]T. Ishihara and H. Yasuura. “Voltage Scheduling Problem for Dynamically Variable Voltage Processors，” International Symposium on Low Power Electronics and Design， pp. 197-202， August 1998.
[16]J. W. S. Liu. Real-Time Systems. Prebtice Hall， 2000.
[17]G. Buttazzo. Hard Real-Time Computing Systems. Kluwer Academic Publishers，
1997.
[18]G. Qu and M. Potkonjak. “Techniques for Energy Minimization of
Communication Pipelines，” IEEE/ACM International Conference on Computer Aided Design， pp. 597-600， November 1998.
[19]G. Quan and X. Hu. “Energy Efficient Fixed-Priority Scheduling for Real-Time Systems on Variable Voltage Processors，” DAC’01: IEEE/ACM Design Automation Conference， pp. 828-833， June 2001.
[20]S. Raje and M. Sarrafzadeh. “Variable voltage scheduling，”ISLPED’95:
International Symposium on Low Power Electronics and Design， pp. 9-14，
April 1995.
[21]F. Yao， A. Demers and S. Shenker. A Scheduling Model for Reduced CPU Energy.
IEEE Annual Foundations of Computer Science， pp. 374 -382， 1995.
[22]I. Hong， M. Potkonjak and M. B. Srivastava. On-line Scheduling of Hard Real-Time Tasks on Variable Voltage Processor. In Computer-Aided Design (ICCAD)’98. pp. 653-656.
[23]I. Hong， D. Kirovski， G. Qu， M. Potkonjak and M. Srivastava. Power optimization of variable voltage core-based systems. In Proceedings of the 35th Design Automation Conference， DAC’98
[24]I. Hong， G. Qu， M. Potkonjak and M. Srivastava. “Synthesis Techniques for Low-Power Hard Real-Time Systems on Variable Voltage Processors.” In Proceedings of 19th IEEE Real-Time Systems Symposium (RTSS’98)， Madrid，
December 1998.
[25]H. Aydin， R. Melhem， D. Moss´e and P.M. Alvarez.”Determining Optimal Processor Speeds for Periodic Real-Time Tasks with Different Power Characteristics.” In Proceedings of the 13th EuroMicro Conference on Real-Time Systems (ECRTS’01)， Delft， Netherlands， June 2001.
[26]H. Aydin， R. Melhem， D. Moss´e and P.M. Alvarez. “Dynamic and Aggressive Scheduling Techniques for Power-Aware Real-Time Systems.” In Proceedings of 22nd Real-Time Systems Symposium， December 2001.
[27]R. Ernst and W. Ye. Embedded Program Timing Analysis based on Path Clustering and Architecture Classification. In Computer-Aided Design (ICCAD)’97. pp.
598-604.
[28]H. Aydin， R. Melhem， D. Moss´e and P.M. Alvarez. Determining Optimal Processor Speeds for Periodic Real-Time Tasks with Different Power Characteristics. In Proceedings of the 13th EuroMicro Conference on Real-Time Systems (ECRTS’01)， Delft， Netherlands， June 2001.
[29]C.L. Liu and J.W.Layland. “Scheduling Algorithms for Multiprogramming in Hard Real-time Environment.” J. of ACM 20(1): pp.46-61， 1973.
[30]Intel， Intel StrongARM SA1110 CPU
http://www.intel.com/design/pca/applicationsprocessors/datashts/index.h tm
[31]D. Grunwald， P. Levis， C. Morrey III， M. Neufeld， and K. Farkas. “Policies
Implementation， Oct 2000.
[32]T. Pering， T. Burd， and R. Brodersen. “Voltage Scheduling in the lpARM Microprocessor System.” International Symposium on Low Power Electronics and Design 2000， pp.96-101， 2000.
[33]http://www.transmeta.com
[34]M. Weiser， B. Welch， A. J. Demers， and S. Shenker. “Scheduling for reduced CPU energy.” In Operating Systems Design and Implementation (OSDI ’94)，
pages 13–23， 1994.
[35]A. Chandrakasan， S. Sheng， and R.W.Brodersen. “Low-power CMOS digitaldesign.” IEEE J. Solid-State Circuits， 27(4):473-484， 1992.
附錄 A 測試參數表
WCET Execution Time Cycles Period
112.5 11.25 2488.5 112.5 562.5 42.1875 62213 562.5 0.25
843.75 63.28125 93319 843.75
225 22.5 4977 225
1125 84.375 62213 1125 0. 5
1687.5 126.5625 93319 1687.5
450 45 9954 450
2250 168.75 62213 2250 1
3375 253.125 93319 3375
900 90 19908 900
4500 337.5 62213 4500 2
6750 506.25 93319 6750
1350 135 29862 1350
6750 506.25 62213 6750 3
10125 759.375 93319 10125
1800 180 39816 1800
9000 675 62213 9000
4
13500 1012.5 93319 13500
2250 225 49770 2250
11250 843.75 62213 11250 5
16875 1265.625 93319 16875
4500 450 99540 4500
22500 1687.5 62213 22500 10
33750 2531.25 93319 33750
6750 675 149310 6750
33750 2531.25 62213 33750 15
50625 3796.875 93319 50625
9000 900 199080 9000
45000 3375 62213 45000 20
67500 5062.5 93319 67500 11250 1125 248850 11250 56250 4218.75 62213 56250 25
84375 6328.125 93319 84375 13500 1350 298620 13500 67500 5062.5 62213 67500 30
101250 7593.75 93319 101250
WCET Execution Time Cycles Period
112.5 28.125 6221 112.5
562.5 70.3125 62213 562.5
0.25
225 56.25 12442.5 225
1125 140.625 62213 1125
1. 5
1687.5 210.9375 93319 1687.5
450 112.5 24885 450
2250 281.25 62213 2250
1
3375 421.875 93319 3375
900 225 49770 900
4500 562.5 62213 4500
2
6750 843.75 93319 6750
1350 337.5 74655 1350
6750 843.75 62213 6750
3
10125 1265.625 93319 10125
1800 450 99540 1800
9000 1125 62213 9000
4
13500 1687.5 93319 13500
2250 562.5 124425 2250
11250 1406.25 62213 11250
5
16875 2109.375 93319 16875
4500 1125 248850 4500
22500 2812.5 62213 22500
10
33750 4218.75 93319 33750
6750 1687.5 373275 6750
33750 4218.75 62213 33750
15
50625 6328.125 93319 50625
9000 2250 497700 9000
45000 5625 62213 45000
20
67500 8437.5 93319 67500
11250 2812.5 622125 11250
56250 7031.25 62213 56250
25
84375 10546.875 93319 84375
13500 3375 746550 13500
67500 8437.5 62213 67500
30
101250 12656.25 93319 101250
WCET Execution Time Cycles Period
112.5 28.125 6221 112.5
562.5 70.3125 62213 562.5
0.25
843.75 105.46875 93319 843.75
225 56.25 12442.5 225
1125 140.625 62213 1125
2. 5
1687.5 210.9375 93319 1687.5
2250 281.25 62213 2250
3375 421.875 93319 3375
900 225 49770 900
4500 562.5 62213 4500
2
6750 843.75 93319 6750
1350 337.5 74655 1350
6750 843.75 62213 6750
3
10125 1265.625 93319 10125
1800 450 99540 1800
9000 1125 62213 9000
4
13500 1687.5 93319 13500
2250 562.5 124425 2250
11250 1406.25 62213 11250
5
16875 2109.375 93319 16875
4500 1125 248850 4500
22500 2812.5 62213 22500
10
33750 4218.75 93319 33750
6750 1687.5 373275 6750
33750 4218.75 62213 33750
15
50625 6328.125 93319 50625
9000 2250 497700 9000
45000 5625 62213 45000
20
67500 8437.5 93319 67500
11250 2812.5 622125 11250
56250 7031.25 62213 56250
25
84375 10546.875 93319 84375
13500 3375 746550 13500
67500 8437.5 62213 67500
30
101250 12656.25 93319 101250
WCET Execution Time Cycles Period
112.5 42.1875 9331.875 112.5
562.5 42.1875 62213 562.5
0.25
843.75 253.125 93319 843.75
225 84.375 18663.75 225
1125 84.375 62213 1125
0.5
1687.5 506.25 93319 1687.5
450 168.75 37327.5 450
2250 168.75 62213 2250
1
3375 1012.5 93319 3375
900 337.5 74655 900
2
6750 2025 93319 6750
1350 506.25 111982.5 1350
6750 506.25 62213 6750
3
10125 3037.5 93319 10125
1800 675 149310 1800
9000 675 62213 9000
4
13500 4050 93319 13500
2250 843.75 186637.5 2250
11250 843.75 62213 11250
5
16875 5062.5 93319 16875
4500 1687.5 373275 4500
22500 1687.5 62213 22500
10
33750 10125 93319 33750
6750 2531.25 559912.5 6750
33750 2531.25 62213 33750
15
50625 15187.5 93319 50625
9000 3375 746550 9000
45000 3375 62213 45000
20
67500 20250 93319 67500
11250 4218.75 933187.5 11250
56250 4218.75 62213 56250
25
84375 25312.5 93319 84375
13500 5062.5 1119825 13500
67500 5062.5 62213 67500
30
101250 30375 93319 101250