This paper surveys the design of embedded computer systems, which use software running on programmable computers to im-plement system functions. Creating an embedded computer system which meets its performance, cost, and design time goals is a hardware-software co-design problewhe design of the hard-ware and software components influence each other. This paper emphasizes a historical approach to show the relationships be-tween well-understood design problems and the as-yet unsolved problems in co-design. We describe the relationship between hard-ware and sofhvare architecture in the early stages of embedded system design. We describe analysis techniques for hardware and software relevant to the architectural choices required for hard-ware-software co-design. We also describe design and synthesis techniques for co-design and related problems.

This paper summarizes the state of the real-time field in the areas of scheduling and operating system kernels. Given the vast amount of work that has been done by both the operations research and computer science communities in the scheduling area, we discuss four paradigms underlying the scheduling approaches and present several exemplars of each. The four paradigms are: static table-driven scheduling, static priority preemptive scheduling, dynamic planning-based scheduling, and dynamic best efSort scheduling. In the operating system context, we argue that most of the proprietary commercial kernels as well as real-time extensions to time-sharing operating system kernels do not fit the needs of predictable real-time systems. We discuss several research kernels that are currently being built to explicitly meet the needs of real-time applications. I.

The ideas and findings in this report should not be construed as an official DoD position. It is published in the interest of scientific and technical information exchange. FOR THE COMMANDER (signature on file)

Next generation real-time systems will require greater.flexibility and pre-dictability than is commonly found in today's systems. These future systems include the space station, integrated vision/robotics/AI systems, collections of humans/robots coordinating to achieve common objectives (usually in haz-ardous environments such as undersea exploration or chemical plants), and various command and control applications. The complexity of such systems due to timing constraints, concurrency, and distribution is high. It is accepted that the synchronization, failure atomicity, and permanence properties of trans-actions aid in the development of distributed systems. However, little work has been done in exploiting transactions in a real-time context. We have been at-tempting to categorize real-time data into classes depending on their time, synchronization, atomicity, and permanence properties. Then, using the se-mantics of the data and the applications, we are developing special, tailored, real--time transactions that only supply the minimal properties necessary for that class. This reduces the system overhead in supporting access to various types of data. The eventual goal is to verify that timing requirements can be met.

The problem of scheduling a set of sporadic tasks that share a set of serially reusable, single unit software resources on a single processor is considered. The correctness conditions are that (1) each invocation of each task completes execution at or before a well-defined deadline, and (2) a resource is never accessed by more than one task simultaneously. We present an optimal on-line algorithm for scheduling a set of sporadic tasks. The algorithm results from the integration of a synchronization scheme for access to shared resources with the earliest deadline first algorithm. A set of relations on task parameters that are necessary and sufficient for a set of tasks to be schedulable is also derived. Our model for the analysis of processor scheduling policies is novel in that it incorporates minimum as well as maximum processing time requirements of tasks. The scheduling algorithm and the sporadic tasking model have been incorporated into an operating system kernel and used to impleme...

: In many real-time applications, the set of tasks in the system as well as the characteristics of the tasks change during system execution. Specifically, the system moves from one mode of execution to another as its mission progresses. A mode change is characterized by the deletion of some tasks, addition of new tasks, or changes in the parameters of certain tasks, e.g., increasing the sampling rate to obtain a more accurate result. This paper discusses a protocol for systematically accomplishing mode change in the context of a priority-driven preemptive scheduling environment. 1. Introduction To successfully develop a large-scale real-time system, we must be able to manage both the logical complexity and timing complexity by using a disciplined approach. The logical complexity is addressed by software engineering methodology, while the timing complexity is addressed by research in real-time scheduling algorithms [2, 3, 5, 6, 7, 8, 10, 12, 14]. An important class of scheduling algori...

This paper concerns the problem of scheduling sets of preemptable, periodic tasks on multiple resources. We consider a task model that allows arbitrary mixes of fixed and migratable tasks, and prove the existence of an optimal pfair scheduler in this model. Fixed tasks must always be scheduled on a given resource, while migratable tasks can be scheduled on different resources at different times. A pfair scheduler produces a periodic schedule in which the times each task is allocated a processor are approximately evenly spread throughout its period. This work extends work of Baruah et al, who proved a similar result for systems in which all tasks are migratable.

Recent results in the application of... this paper. The review takes the form of an analysis of the problems presented by different application requirements and characteristics. Issues covered include uniprocessor and multiprocessor systems, periodic and aperiodic processes, static and dynamic algorithms, transient overloads and resource usage. Protocols that limit and reduce blocking are discussed. Considerations are also given to scheduling Ada tasks.

We consider the problem of scheduling a set of n tasks in a system having r resources. Each task has an arbitrary, but known, processing time and a deadline, and may request use of a number of resources. A resource can be used either in shared mode or exclusive mode. In this article, we study algorithms used for determining whether or not a set of tasks is schedulable in such a system, and if so, determining a schedule for it. This scheduling problem is known to be NPcomplete and hence we methodically study a set of heuristics that can be used by such an algorithm. Due to the complex-ity of the problem, simple heuristics do not perform satisfactorily. However, an algorithm that uses combina-tions of these simple heuristics works very well com-pared to an optimal algorithm that takes exponential time complexity. For the combination that performs the best, we also determine the scheduling costs as a function of the size of the task set scheduled. 1.

Recent results in the application of scheduling theory to dependable real-time systems are reviewed. The review takes the form of an analysis of the problems presented by different application requirements and characteristics. Issues covered include uniprocessor and multiprocessor systems, periodic and aperiodic processes, static and dynamic algorithms, transient overloads and resource usage. Protocols that bound and reduce blocking are discussed. A review of specific real-time kernels is also included.