Energy Management of Virtual Memory on Diskless Devices

Abstract

In a pervasive computing environment, applications are able to run across different platforms with significantly different resources. Such platforms range from highperformance desktops to handheld PDAs. This paper discusses a compiler approach to reduce the energy consumption of a diskless device where the swap space is provided by a remotely mounted file system accessible via a wireless connection. Predicting swapping events at compile time allows effective energy management of a PDAs wireless communication component such as a 802.11 or Bluetooth card.The compiler activates and deactivates the communication card based on compile-time knowledge of the past and future memory footprint of an application. In contrast to OS techniques, the compiler can better predict future program behavior, and can change this behavior through program transformations that enable additional optimizations.A prototype compilation system EEL RM has been implemented as part of the SUIF2 compiler infrastructure. Preliminary experiments based on the SimpleScalar simulation toolset and three numerical programs indicate the potential benefits of the new technique.

Energy characterization of filesystems for diskless embedded systems

“…A recent study of energy management of virtual memory for diskless devices has been carried out in [1]. This work concerns energy management for the main memory and does not consider secondary storage.…”

Inter-program optimisations for disk energy reduction

“…An energy-aware compiler can reshape a program such that the idle times between successive resource accesses are maximized, giving opportunities to hibernate a device more often, and/or in deeper hibernation states. This compilation strategy has been shown to work well in a single process environment [2][3][4], but may lead to poor overall results in a multiprogramming environment. In a multiprogramming setting, one program may finish accessing a resource and may direct the resource to hibernate during some time of idleness.…”

Inter-program optimizations for conserving disk energy

“…This compilation strategy has been shown to work well in a single process environment [5,4,6], but may lead to poor overall results in a multiprogramming environment. In a multiprogramming setting, one program may finish accessing a resource and then direct the resource to hibernate over its idle period.…”