Abstract:
Over the past few years there has been a trend in the embedded software
development community to move software development to more advanced
object-oriented languages and runtimes. The movement is occasioned by a
systemic increase in the complexity and size of the software required by
commercially successful embedded devices. One of the distinguishing
characteristics of embedded software is that it requires predictable
execution of logic, i.e., it is real-time. A spacecraft is an
interesting example of an embedded system. Indeed, NASA/Caltech's Jet
Propulsion Laboratories faces significant software engineering
challenges due to (1) the complexity of planned missions, (2) the lack
of an overarching software architecture for spacecraft, and (3) the lack
of advanced real-time programming languages and runtimes. The Golden
Gate project, Sun Microsystems Laboratories, JPL, and Carnegie Mellon
University, will attempt to implement a software architecture developed
at JPL for spacecraft (the Mission Data System (MDS)) using the
Real-Time Specification for Java (an advanced real-time software
development language and runtime). This talk explores the challenges and
goals of the Golden Gate project.
Speaker Bio:
Dr. Greg Bollella, Principal Investigator for Real-Time Java at Sun
Microsystems Laboratories , has been interested in algorithms and
software architectures that support the deterministic execution of logic
within general-purpose operating systems and virtual machines since
1992. While a Senior Architect at IBM, he led the Real-Time for Java
Expert Group under the Java Community Process which developed the
Real-Time Specification for Java (RTSJ). At Sun Labs, he continues to be
actively involved in internal and external projects which support the
RTSJ, especially in architectures for the emerging sensor industry. He
concurrently holds a position at NASA's Jet Propulsion Laboratories as
Distinguished Visiting Scientist and is working on developing software
architectures for spacecraft. Greg holds a Ph.D. in computer science
from the University of North Carolina at Chapel Hill. His dissertation
research is in real-time scheduling theory and real-time systems
implementation.