GRACE: A Hierarchical Adaptation Framework for Saving Energy

Mobile systems primarily processing multimedia data are expected to become important platforms for pervasive computing. These systems, however, must satisfy large, dynamic demands of multimedia applications subject to stringent energy, computational, and bandwidth constraints. At the same time, multimedia applications provide the possibility of adaptation, allowing tradeoffs between energy, computation, and network bandwidth to maximize the user’s experience for the current resources. Researchers have proposed adaptations in the hardware, network, operating system, and applications to provide QoS guarantees and to save energy. To reap the full benefits of such adaptations, however, the different system layers and applications must coordinate their adaptations with each other. This paper describes a framework, called GRACE, to achieve such a coordination, using a novel hierarchical approach that combines global, per-application, and per-layer internal adaptation, for multimedia applications running on wireless systems. GRACE achieves the benefits of coordination through cleanly defined interfaces that keep the internals of the different layers isolated from each other. Our results so far show the effectiveness of the hierarchical adaptations, and justify the use of coordinated, cross-layer adaptations to both save energy and improve the user’s multimedia experience.

[1]  Douglas L. Jones,et al.  Cross-layer adaptive video coding to reduce energy on general-purpose processors , 2003, Proceedings 2003 International Conference on Image Processing (Cat. No.03CH37429).

[2]  Scott Shenker,et al.  Scheduling for reduced CPU energy , 1994, OSDI '94.

[3]  Karsten Schwan,et al.  Power-aware communication for mobile computers , 1999, 1999 IEEE International Workshop on Mobile Multimedia Communications (MoMuC'99) (Cat. No.99EX384).

[4]  Amin Vahdat,et al.  ECOSystem: managing energy as a first class operating system resource , 2002, ASPLOS X.

[5]  Eyal de Lara,et al.  Reducing the Energy Usage of Office Applications , 2001, Middleware.

[6]  Mark D. Corner,et al.  Fugue: time scales of adaptation in mobile video , 2000, IS&T/SPIE Electronic Imaging.

[7]  Christian Poellabauer,et al.  Cooperative run-time management of adaptive applications and distributed resources , 2002, MULTIMEDIA '02.

[8]  Christopher J. Hughes,et al.  Joint local and global hardware adaptations for energy , 2002, ASPLOS X.

[9]  Willy Zwaenepoel,et al.  HATS: hierarchical adaptive transmission scheduling for multi-application adaptation , 2001, IS&T/SPIE Electronic Imaging.

[10]  Keith Cheverst,et al.  A platform supporting coordinated adaptation in mobile systems , 2002, Proceedings Fourth IEEE Workshop on Mobile Computing Systems and Applications.

[11]  Thomas Kunz,et al.  Image transcoding for wireless WWW access: the user perspective , 2001, IS&T/SPIE Electronic Imaging.

[12]  Daniel P. Siewiorek,et al.  A scalable solution to the multi-resource QoS problem , 1999, Proceedings 20th IEEE Real-Time Systems Symposium (Cat. No.99CB37054).

[13]  Mahadev Satyanarayanan,et al.  Agile application-aware adaptation for mobility , 1997, SOSP.

[14]  Rami G. Melhem,et al.  Maximizing the system value while satisfying time and energy constraints , 2003, IBM J. Res. Dev..

[15]  Scott A. Brandt,et al.  Flexible Soft Real-Time Processing in Middleware , 2004, Real-Time Systems.

[16]  Douglas L. Jones,et al.  Design and evaluation of a cross-layer adaptation framework for mobile multimedia systems , 2003, IS&T/SPIE Electronic Imaging.

[17]  Rohit Jain,et al.  Variability in the execution of multimedia applications and implications for architecture , 2001, ISCA 2001.