CPU scheduling for statistically-assured real-time performance and improved energy efficiency

We present a CPU scheduling algorithm, called Energy-efficient Utility Accrual Algorithm (or EUA), for battery-powered, embedded real-time systems. We consider an embedded software application model where repeatedly occurring application activities are subject to deadline constraints specified using...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:CODES+ISSS 2004 : International Conference on Hardware/Software Codesign and System Synthesis : September 8-10, 2004, Stockholm, Sweden s. 110 - 115
Hlavní autoři: Wu, Haisang, Ravindran, Binoy, Jensen, E. Douglas, Li, Peng
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: New York, NY, USA ACM 08.09.2004
IEEE
Edice:ACM Conferences
Témata:
Application software
Applied computing > Computers in other domains
Computer systems organization > Embedded and cyber-physical systems
Computer systems organization > Embedded and cyber-physical systems > Embedded systems
Computer systems organization > Real-time systems
Costs
Embedded software
Embedded system
Energy consumption
Energy efficiency
Polynomials
Processor scheduling
Real time systems
Scheduling algorithm
Software and its engineering > Software organization and properties > Contextual software domains > Operating systems > Process management
Software and its engineering > Software organization and properties > Software system structures > Embedded software
Software and its engineering > Software organization and properties > Software system structures > Real-time systems software
energy-efficient scheduling
real-time systems
time/utility functions
utility accrual scheduling
ISBN:1581139373, 9781581139372
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:We present a CPU scheduling algorithm, called Energy-efficient Utility Accrual Algorithm (or EUA), for battery-powered, embedded real-time systems. We consider an embedded software application model where repeatedly occurring application activities are subject to deadline constraints specified using step time/utility functions. For battery-powered embedded systems, system-level energy consumption is also a primary concern. We consider CPU scheduling that (1) provides assurances on individual and collective application timeliness behaviors and (2) maximizes system-level timeliness and energy efficiency. Since the scheduling problem is intractable, EUA heuristically computes CPU schedules with a polynomial-time cost. Several properties of EUA are analytically established, including timeliness optimality during under-load situations and statistical assurances on timeliness behavior. Further, our simulation results confirm EUA's superior performance.
Bibliografie:SourceType-Conference Papers & Proceedings-1
ObjectType-Conference Paper-1
content type line 25
ISBN:1581139373
9781581139372
DOI:10.1145/1016720.1016749