This object-based, real-time fault tolerance scheme is called the primary-shadow RTO.k replication (PSRR) scheme and has been validated through several real-time computing application development experiments conducted in the authors' laboratory.
In an effort to validate the PSRR scheme, several experimental developments of fault-tolerant real-time computing application systems have been made in the authors' laboratory.
In Proceedings of the International Workshop on Real-Time Computing Systems and Applications (RTCSA) (Tokyo, Japan, Oct.
While all the challenges and important research areas (see Lee and Krishna [1993]; Son [1995]; Stankovic [1988]; Stankovic and Ramamritham [1988]) in real-time computing cannot be covered here, several key areas are stressed.
Nevertheless, existing real-time computing infrastructure often introduces formidable barriers to continuous process improvement, equipment upgrades, and agility in responding to changing markets and increased global competition.
For example, in response to such problems, Chrysler, Ford, and GM have worked together and stated their requirements for next-generation real-time computing systems in a white paper, "Requirements of Open, Modular Architecture Controllers for Applications in the Automotive Industry" (Aug.
Other issues addressed included scheduling and resource management, system design, software design, wireless and sensor networks, industrial advances, middleware, systems modeling, modeling with Unified Modeling Language, real-time systems, fundamental issues in distributed
real-time computing, language support and design patterns, data communication, fault-tolerance and security, and performance evaluation.
Researchers from academia and industry describe recent work in the field of embedded and
real-time computing. The 37 contributions address such topics as analyzing time division multiple access networks with slot skipping; load balancing in bounded-latency content distribution; and allocating data to scratchpad memory in a way that minimizes a task's worst-case execution time.
From an August 2005 international conference on embedded and
real-time computing systems and applications organized by Hong Kong Baptist University's Department of Computer Science come 105 papers originally presented in sessions dealing with scheduling, operating systems, embedded systems, communications, QoS (quality-of-service) support, sensor networks and information systems, energy aware design, data services, case studies and industrial experience, distributed/multiprocessor systems, mobile and ubiquitous computing, and hardware/software codesign.
Papers examine issues related to embedded
real-time computing, industry perspectives, smart spaces, mission-critical computing, and service, application, and device integration and implementation.
In these proceedings from the July 2005 conference, contributors describe their work in energy-aware computing (including speed modulation), worst-case execution time analysis (including a WCET- oriented branch prediction system), programming languages, modeling and validation techniques including model-based and component-based approaches, operating system support, scheduling and "schedulability" analysis, quality-of-service support and wireless sensor networks (including scheduling task with Markov-chain-based constraints), multiprocessor systems, and applications of
real-time computing.