Next Generation Software: Scalable Real-Time Simulation of Embedded Systems and Environments

点击翻译按钮获取中文摘要

Scalable Distributed Real-Time Simulation of Embedded Systems and Environments Kane Kim University of California -Irvine 0204050 Major improvements in the validation technology for embedded systems are under increasing demands from industry. Not only description but also simulation of non-computer parts and application environments of embedded systems is needed in validating many embedded system software designs and implementations. Here the desired types of simulators are real-time (RT) simulators which exhibit the timing behavior that are the same as or sufficiently close to the timing behavior of the simulation targets. Such simulators can enable highly cost-effective testing of the embedded system software and such testing can be a lot cheaper that the testing performed in actual application environments while being much more effective than the testing based on non-RT simulators of environments. In this research, the joint university-industry research team intends to establish the scientific foundation for the DTS scheme in a sound form, thereby realizing fundamental advances in the state of the art in RT simulation. Development of support middleware and other tool prototypes is planned. Successful accomplishment of these will mean that the validation technology for embedded systems will be advanced in fundamental ways. Case studies for evaluating the improvements in simulation performance and validation effectiveness realized by use of the DTS scheme in practical contexts will be conducted with the assistance of industry partners. Advanced instrumentation and performance measurement technologies and graphic interface technologies developed by international collaborations partners will also be incorporated and integrated with the DTS scheme. More specifically, the proposing team intends to establish:(1) Scientific foundation for DTS, including that related to consistency among distributed simulator nodes and maximization of concurrency, (2) Middleware and application programming interfaces that support DTS and simulator programming, (3) A methodology for validation of RT object-structured distributed software for embedded systems, which is centered around the use of RT simulators of the non-computer parts and environments and additionally uses performance measurement and graphic visualization tools, and (4) Case studies involving an RT simulation of immune buildings designed to protect inhabitants against bio-weapon attacks and that of miliary command-control environments.