Agile Application-Aware Adaptation for Mobility
Brian D. Noble, M. Satyanarayanan, Dushyanth Narayanan, James Eric Tilton, Jason Flinn, Kevin R. Walker
School of Computer Science
Carnegie Mellon University
In this paper we show that
collaborative partnership between the operating system and
applications, offers the most general and effective approach
to mobile information access. We describe the design of
a prototype implementing this approach, and show
how it supports concurrent execution of diverse mobile ap-
as a key attribute of adap-
tive systems, and describe how to quantify and measure it.
We present the results of our evaluation of Odyssey, indi-
cating performance improvements up to a factor of 5 on a
benchmark of three applications concurrently using remote
services over a network with highly variable bandwidth.
Adaptation is the key to mobility.
Only through alertness and
prompt reactions can a mobile client offer acceptable service in
spite of the many problems that plague its existence. These in-
clude unpredictable variation in network quality, wide disparity in
the availability of remote services, limitations on local resources
imposed by weight and size constraints, concern for battery power
consumption, and lowered trust and robustness resulting from ex-
posure and motion [S, 15,311.
Once the need for adaptation is recognized, many questions fol-
low. What form should such adaptation take? Which system com-
ponents should bear responsibility for adaptation? How does one
characterize the adaptive capability of a mobile client? How does
one compare alternative designs from the perspective of adaptation?
We present our answers to these and related questions in this pa-
per. We describe the design and implementation of a software plat-
and show how it provides effective support
for concurrent execution of diverse mobile applications. We iden-
as a key attribute of adaptive systems, and describe how
to quantify and measure it. Finally, we present the results of our
evaluation of the Odyssey prototype to confirm the benefits of our
approach. These results indicate performance improvementsup to a
factor of 5 on a benchmark of three applications concurrently using
remote services over a network with highly variable bandwidth.
This research was supported by the Air Force Materiel Command (AFMC) under
DARPA contract numbers F19628-93-GO193 and F19628-96GOO61. Additional
support was providedby AT&T, Hughes, IBM, and Intel. The views and conclusions
contained here are those of the authors and should not be interpreted as necessarily
representingthe official policies or endorsements,either express or implied. of AFMC,
DARPA, AT&T, Hughes, IBM, Intel. Carnegie Mellon University, or theU.S. Govem-