Download:
by Heng Zeng, Carla S. Ellis, Alvin R. Lebeck, Amin Vahdat
http://www.cs.duke.edu/ari/millywatt/ecosys-policies.pdf
Add To MetaCart
Abstract:
The global system nature of energy creates challenges in developing operating system policies to effectively manage energy consumption. Our proposed currentcy model creates the framework for the operating system to manage energy as a first-class resource. Furthermore, currentcy provides a powerful mechanism to formulate unified energy management policies across diverse competing applications and spanning devices with very different power characteristics. We claim that unifying energy management enables more coherent and efficient energy consumption. This paper presents an initial exploration of the policy space enabled by the currentcy model as implemented in ECOSystem, our Linux-based prototype. We use ECOSystem to attack four specific energyrelated goals: 1) currentcy conserving scheduling algorithms that reduce residual battery capacity, 2) proportional energy sharing, 3) response time variation, and 4) energy efficient disk management. Our results show that the currentcy model is a powerful framework for expressing better energy management policies than those that come from the traditional perdevice approach. 1
Citations
|
342
|
Resource Containers: A New Facility for Resource Management in Server Systems
– Banga, Druschel, et al.
- 1999
|
|
336
|
Scheduling for reduced CPU energy
– Weiser, Welch, et al.
- 1994
|
|
242
|
Real-time dynamic voltage scaling for low-power embedded operating systems
– Pillai, Shin
- 2001
|
|
233
|
Measuring and reducing energy consumption of network interfaces in handheld devices
– Stemm, Katz
- 1997
|
|
178
|
Energy-Aware Adaptation for Mobile Applications
– Flinn, Satyanarayanan
- 1999
|
|
155
|
Adaptive disk spin-down policies for mobile computers
– Douglis, Krishnan
- 1995
|
|
145
|
Thwarting the power-hungry disk
– Douglis, Krishnan, et al.
- 1994
|
|
137
|
A quantitative analysis of disk drive power management in portable computers
– Li, Kumpf, et al.
- 1994
|
|
131
|
A dynamic disk spin-down technique for mobile computing
– Helmbold, Long, et al.
- 1996
|
|
124
|
Lottery and stride scheduling: Flexible proportionalshare resource management
– Waldspurger
- 1995
|
|
109
|
Policies for dynamic clock scheduling
– Grunwald, Levis, et al.
- 2000
|
|
107
|
Power management techniques for mobile communication
– Kravets, Krishnan
- 1998
|
|
94
|
Power-Aware Page Allocation
– Lebeck, Fan, et al.
- 2000
|
|
83
|
The Case for Higher Level Power Management
– Ellis
- 1999
|
|
79
|
ECOSystem: Managing Energy as a First Class Operating System Resource,” Proc. Int’l Conf. Architectural Support for Programming Languages and Operating Systems (ASPLOS 2002), Oct. 2002. Hakan Aydin received the BS and MS degrees in control and computer en
– Zeng, Fan, et al.
- 1996
|
|
65
|
Adaptive Disk Spindown Via Optimal Rent-to-buy
– Krishnan, Long, et al.
- 1995
|
|
56
|
Every Joule is Precious: A Case for Revisiting Operating System Design for Energy Efficiency
– Vahdat, Lebeck, et al.
- 2000
|
|
33
|
DRAM Energy Management Using Software and Hardware Directed Power Mode Control
– Delaluz, Kandemir, et al.
- 2001
|
|
7
|
The Simulation and Evaluation of Dynamic Scaling Algorithms
– Pering, Burd, et al.
- 1998
|