This paper proposes a divide-and-conquer approach for global throughput optimization which not only leverages upon existing techniques, but enables their more effective and coordinated use. The "divide" approach consists of logical partitioning of the computation into subparts falling into one of a set of preclassified computation types. The subparts are then "conquered" through coordinated application of existing optimization techniques. We have characterized a set of techniques in terms of their expected effect on throughput, and can thus select the most promising techniques for each unique situation. The technique is not limited to a specific class of computations and gives higher, or at worst equal, improvement than previously reported techniques on all examples. 1.0 Introduction Throughput optimization techniques remain important for meeting the sampling rate requirements of modern DSP and communication applications. Though clock rates for ASICs and general purpose computing devi...

This paper presents an overview of NASA Langley's research program in formal methods. The major goals of this work are to make formal methods practical for use on high integrity systems, to orchestrate the transfer of this technology to U.S. industry through use of carefully designed demonstration projects, and to exploit this technology to help achieve NASA's new ambitious goals in aeronautics. Several direct technology transfer efforts have been initiated that apply formal methods to critical subsystems of real aerospace computer systems. The research team consists of six NASA civil servants and contractors from Odyssey Research Associates and SRI International. Contents 1 Rationale For a Formal Methods Research Program 3 1.1 The Problem With Software and Hardware . . . . . . . . . . . . . . . . . . . . . . . 3 1.2 What is Formal Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 2 Goals of Our Program, Strategy, and Research Team 6 2.1 Technology Trans...

ion is a significant tool in managing the software development process. This has long been recognized by the real-time community, as evidenced by the number of real-time kernels available. The kernel allows the programmer to ignore the processor-specific details of task management: allocation, switching, synchronization, scheduling, etc. These kernels have varying features, depending upon the goals of the project. A major distinction between kernels can be drawn in the support for single vs. multiple processor designs. Uniprocessor support is by far the most common. These kernels are acceptable where the processing demands are modest enough to be handled comfortably by a single processor. This technology is well developed, and has been used in the commercial sector for many years. Multiprocessor kernels, such as Spring [37], MARS [36], and Chimera II [38] are at a much more primitive level. These kernels begin to examine the issues of integrating parallel processors into real-time sy...

A critical aspect of developing hypermedia applications is being able to identify the interlinking within the information, and structure it in such a way that enhances accessibility. This is a major part of the process which is referred to as authoring. There are many approaches to authoring. Some are suitable for small prototypes. When developing large hypermedia systems, or systems which are likely to require maintenance over a period of time, it becomes essential to select or more structured approach, and appropriate methodologies to support the development process. The use of an inappropriate approach to authoring can result in systems which have a greatly inflated development cost, which do not provide the level of functionality desired, and are difficult to use and maintain. INTRODUCTION There has recently been a strong focus on the burgeoning information revolution. As a number of enabling technologies (such as communications networks, databases, and computing hardware) continu...

At present the cost of producing large hypermedia information systems is very high and the process is time consuming. Also the way related information is linked together is prone to errors resulting in lower quality applications. The process of creating hypermedia information systems is known as authoring. The way authoring is performed today is analogous to how cars were produced before the introduction of mass production techniques and assembly lines when every part had to be handcrafted. Similarly today many authoring tools require the author to handcraft every screen and manually find and link related information together. This makes these techniques unsuitable for production of large hypermedia information systems. Thus we need a new generation of authoring tools that can support rapid production of higher quality large hypermedia applications in a cost effective way. To get a deeper understanding of the issues that need to be addressed when developing next generation a...