Title of Paper: Adaptive Management of Computing and Network
Resources for Real-Time Sensor Webs
Principal Author: Dr. Lonnie Welch
Abstract: In order to transition from large,
instrument-jammed observatories to cheaper and lighter space-based platforms,
NASA's Earth Science Vision Initiative has formulated the notion of a
sensor-web, a closely integrated constellation of earth observing satellites
that collectively monitor the conditions of our planet through a vast array of
instruments. The network of satellites will act autonomously in controlling
instruments and spacecraft, while also responding to the commands of the user
interested in measuring specific events and to important terrestrial events. It
is likely that NASA's future spacecraft systems will consist of distributed
processes which will handle dynamically varying workloads in response to
perceived scientific events, the spacecraft environment, spacecraft anomalies
and user commands. Since all situations and possible uses of sensors cannot be
anticipated during pre-deployment phases, an approach for dynamically adapting
the allocation of distributed computational and communication resources is
needed. Even if it were possible to anticipate all possible use case scenarios
of future sensor webs, dedicating computational and communication resources
based on maximum requirements will be far more expensive than employing an
adaptive resource management system. To address these issues, we are evolving
the DeSiDeRaTa resource management middleware to accommodate (1) the dynamic
aspects of intra-constellation network topologies and (2) the complete
real-time path from the instrument to the user.
Future autonomous satellites will perform much of the event detection and
response processing that is presently performed by ground-based systems. They
will perform on-board processing of sensor data through the use of high
performance processing architectures and reconfigurable computing environments.
Thus, we will demonstrate the feasibility of adaptive resource management for a
ground-based command and control system currently used by several NASA
satellite systems. In addition to showing feasibility, this will provide a
ground-based testbed that will enable NASA to perform early evaluation of
sensor web management techniques.