|
|
Networked Real Time and Embedded Systems Laboratory
Department of
Computer Science
The
|
|||||
|
|
||||||
Resource
Management
|
Integration
|
Dependency Algebra
|
Virtualization
|
FPGAs
|
Overview
of System Integration Research
Cyber Physical Systems represent the convergence of
computing, communication and intelligence sensing and control of our physical
environments. They are at the center of modern society’s computer
controlled and networked physical infrastructures, including networked
medical devices, transportation systems, defense systems and future tele-presence systems.
System integration uses not only new components
developed specifically for the current requirements but also a large number
of existing components designed under different assumptions and using
different real time, fault tolerance and security protocols. They also have a
different degree of reliability. From a system
engineering perspective, the key challenges include:
·
Unexpected interactions resulting from
syntactically compatible but semantically inconsistent QoS
protocols used by different components; from implicit and outdated
assumptions about the environment embedded in the old components; and from
incompatible real time, fault tolerance, and security protocols.
·
System instabilities when faults and failures
in one component cascade along complex and unexpected interdependency
relations, leading to system wide failures.
·
Inadequate system integration infrastructure due to the lack of
integrated and reusable domain specific reference architectures, tools, and
design patterns with known real time, robustness, and security properties.
Research Areas
·
Predictable composition:
machine checkable annotations of component QoS
protocol usage and semantic properties; explicit and machine checkable component
assumptions about external environments; sets of formally specified and verified coherent
real time, robustness, security and networking protocols.
·
Robust real time software architecture: simple and analytically redundant service for
robustness against software and hardware faults and failures; networked fault
containers for provably safe sharing of computing and networking resources; technologies and tools for
dependency reduction and tracking; hardware
and software co-scheduling technology for predictable timing behaviors
·
System integration infrastructure: integrated sets of domain
models, reference architectures, design patterns and QoS
protocols for interaction complexity reduction and for predictable real time,
robustness, and security properties.