SEARCH · Search NASA
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
SDS-930 computer determines impact cost-decision value at any chronological point in definition phase of multiprogram project. Options, option combinations, and different displays are listed. Graphical procedure was chosen as best method for project.
The computer executive functional system design concepts derived from study of the Space Station/Base are presented. Information Management System hardware configuration as directly influencing the executive design is reviewed. The hardware configuration and generic executive design requirements are considered in detail in a previous report (System Configuration and Executive Requirements Specifications for Reusable Shuttle and Space Station/Base, 9/25/70). This report defines basic system primitives and delineates processes and process control. Supervisor states are considered for describing basic multiprogramming and multiprocessing systems. A high-level computer executive including control of scheduling, allocation of resources, system interactions, and real-time supervisory functions is defined. The description is oriented to provide a baseline for a functional simulation of the computer executive system.
In a preliminary investigation into the feasibility of establishing a data center for earth physics, 12 disciplines were considered for inclusion. Estimation of the size of the data base for each indicated a need for storage of approximately 10 to the 10th power characters. The computer-based system deemed most worthy of further investigation was the interactive concept with remote-terminal access. Users are divided into three classes according to how they would access information: with no terminal, with an interactive terminal, and with a multidevice terminal. All these users can be served by the same center without any particular difficulty, but the real benefactor is the user with an interactive terminal, because he can compile, debug, and run programs in one continuous session. Final points stressed are multiprogramming for dynamic resource sharing, hardware modularity for future expansion, and information protection for such a large community of users. It is concluded that a survey should be conducted to gather more information from the potential users of such a system, and that a pilot project should be developed at some location where both earth-physics research and data-processing capabilities already exist.
Discussion of an approach for implementing digital filters using microprogrammed control logic with read only memory (ROM) for strapdown guidance applications, and description of a second-order multiplexed system using multiprogrammed control instruction. The microprogramming technique for control using ROM is shown to enhance higher order digital filter realization. The high speed of digital circuits reduces propagation time and facilitates multiplexing. Programmed ROM's can be altered easily for different algorithms.
Higher order software (HOS) is concerned only with computable functions and relationships. The HOS methodology can be used for the definition of software for multiprogrammed, multiprocessor, or multicomputer systems. A description of HOS methodology is presented, giving attention to questions of formulation, interface correctness, specification language principles, and HOS analyzers. Aspects of system design are considered, and details of software management are discussed. Attention is given to modularity as defined by HOS, frozen module management, the assembly control supervisor, and aspects of reliability and efficiency.
A processing system is presented that implements simultaneously the efficiency of the special-purpose processor and the total applicability of the general-purpose computer - characteristics commonly thought of as being mutually exclusive. This is achieved through specializing the machine by programming the hardware structure, rather than by adding software systems to it. Data are organized in circulating pages which form a multiplicity of local dynamic memories for each process. Programs are made up of modules, each describing a transient special-purpose machine. A characteristic of this approach is that the processes are data-driven, rather than program-driven. The programming language presents significant flexibility and efficiency in modeling certain classes of problems, and it may be of interest as an implementation model in a broader context. Applications to real-time processing of radar signals are reported. The relevance of characteristics of this system to problems in multiprogramming and multiprocessing systems is discussed.
Queuing network models of a computer system operating with a single workload type are presented. Program algorithms are adapted for use on the Texas Instruments SR-52 programmable calculator. By slightly altering the algorithm to process the G and H matrices row by row instead of column by column, six devices and an unlimited job/terminal population could be handled on the SR-52. Techniques are also introduced for handling a simple load dependent server and for studying interactive systems with fixed multiprogramming limits.
Test circuit times routines even during multiprogram operation. Circuit generates pulse started by signal at beginning address of program under test and ended by signal at ending address. Pulse duration measured with logic analyzer to determine execution time.
Real-time expert systems for monitoring and control are driven by input data which changes with time. One of the subtle problems of this field is the propagation of time variant problems from rule to rule. This propagation problem is even complicated under a multiprogramming environment where the expert system may issue test commands to the system to get data and to access time consuming devices to retrieve data for concurrent reasoning. Two approaches are used to handle the flood of input data. Snapshots can be taken to freeze the system from time to time. The expert system treats the system as a stationary one and traces changes by comparing consecutive snapshots. In the other approach, when an input is available, the rules associated with it are evaluated. For both approaches, if the premise condition of a fired rule is changed to being false, the downstream rules should be deactivated. If the status change is due to disappearance of a transient problem, actions taken by the fired downstream rules which are no longer true may need to be undone. If a downstream rule is being evaluated, it should not be fired. Three mechanisms for solving this problem are discussed: tracing, backward checking, and censor setting. In the forward tracing mechanism, when the premise conditions of a fired rule become false, the premise conditions of downstream rules which have been fired or are being evaluated due to the firing of that rule are reevaluated. A tree with its root at the rule being deactivated is traversed. In the backward checking mechanism, when a rule is being fired, the expert system checks back on the premise conditions of the upstream rules that result in evaluation of the rule to see whether it should be fired. The root of the tree being traversed is the rule being fired. In the censor setting mechanism, when a rule is to be evaluated, a censor is constructed based on the premise conditions of the upstream rules and the censor is evaluated just before the rule is fired. Unlike the backward checking mechanism, this one does not search the upstream rules. This paper explores the details of implementation of the three mechanisms.
The NASA Johnson Space Center's new Multiprogram Control Center (MPCC) addresses the control requirements of complex STS payloads as well as unmanned vehicles. An account is given of the relationship of the MPCC to the STS Mission Control Center, with a view to significant difficulties that may be encountered and solutions thus far devised for generic problems. Examples of MPCC workstation applications encompass telemetry decommutation, engineering unit conversion, data-base management, trajectory processing, and flight design.
The Mission Control Center (MCC) is the facility that provides centralization of all mission operations in support of the shuttle from liftoff through landing. This facility serves as the focal point for real time flight control operations supporting the shuttle, spacelab, and selected payload systems. MCC upgrade accomplishments are as follows: replaced 5 mainframe computers and the associated input/output data interfaces; developed and installed a distributed data system with 81 workstations, 5 LANs, data drivers, and a fiber optic distribution subsystem; developed over 1.5 million lines of code to support new distributed system; developed new digital voice system which provides the MCC capacity to handle th Shuttle programs demands and growth potential to handle Space Station; replaced failing digital TV equipment with digital generation equipment; replaced obsolete center screen projector; and provided the platform for the multiprogram control center.
The Data Management System (DMS) Advanced Development task provides for the development of concepts, new tools, DMS services, and for the testing of the Space Station DMS hardware and software. It also provides for the development of techniques capable of determining the effects of system changes/enhancements, additions of new technology, and/or hardware and software growth on system performance. This paper will address the built-in characteristics which will support network monitoring requirements in the design of the evolving DMS network implementation, functional and performance requirements for a real-time, multiprogramming, multiprocessor operating system, and the possible use of advanced development techniques such as expert systems and artificial intelligence tools in the DMS design.
Effective load-balancing policies use dynamic resource information to schedule tasks in a distributed computer system. We present a novel method for automatically learning such policies. At each site in our system, we use a comparator neural network to predict the relative speedup of an incoming task using only the resource-utilization patterns obtained prior to the task's arrival. Outputs of these comparator networks are broadcast periodically over the distributed system, and the resource schedulers at each site use these values to determine the best site for executing an incoming task. The delays incurred in propagating workload information and tasks from one site to another, as well as the dynamic and unpredictable nature of workloads in multiprogrammed multiprocessors, may cause the workload pattern at the time of execution to differ from patterns prevailing at the times of load-index computation and decision making. Our load-balancing policy accommodates this uncertainty by using certain tunable parameters. We present a population-based machine-learning algorithm that adjusts these parameters in order to achieve high average speedups with respect to local execution. Our results show that our load-balancing policy, when combined with the comparator neural network for workload characterization, is effective in exploiting idle resources in a distributed computer system.
A team was established which consisted of representatives from NASA (Marshall Space Flight Center and Langley Research Center), Thiokol Corporation, the University of Alabama in Huntsville, AC Engineering, SAIC, Martin Marietta, and Aerojet. The team's purpose was to bring together the appropriate personnel to determine what surface inspection techniques were applicable to multiprogram bonding surface cleanliness inspection. In order to identify appropriate techniques and their sensitivity to various contaminant families, calibration standards were developed. Producing standards included development of consistent low level contamination application techniques. Oxidation was also considered for effect on inspection equipment response. Ellipsometry was used for oxidation characterization. Verification testing was then accomplished to show that selected inspection techniques could detect subject contaminants at levels found to be detrimental to critical bond systems of interest. Once feasibility of identified techniques was shown, selected techniques and instrumentation could then be incorporated into a multipurpose inspection head and integrated with a robot for critical surface inspection. Inspection techniques currently being evaluated include optically stimulated electron emission (OSEE); near infrared (NIR) spectroscopy utilizing fiber optics; Fourier transform infrared (FTIR) spectroscopy; and ultraviolet (UV) fluorescence. Current plans are to demonstrate an integrated system in MSFC's Productivity Enhancement Complex within five years from initiation of this effort in 1992 assuming appropriate funding levels are maintained. This paper gives an overview of work accomplished by the team and future plans.
Many scientific applications have intense computational and I/O requirements. Although multiprocessors have permitted astounding increases in computational performance, the formidable I/O needs of these applications cannot be met by current multiprocessors a their I/O subsystems. To prevent I/O subsystems from forever bottlenecking multiprocessors and limiting the range of feasible applications, new I/O subsystems must be designed. The successful design of computer systems (both hardware and software) depends on a thorough understanding of their intended use. A system designer optimizes the policies and mechanisms for the cases expected to most common in the user's workload. In the case of multiprocessor file systems, however, designers have been forced to build file systems based only on speculation about how they would be used, extrapolating from file-system characterizations of general-purpose workloads on uniprocessor and distributed systems or scientific workloads on vector supercomputers (see sidebar on related work). To help these system designers, in June 1993 we began the Charisma Project, so named because the project sought to characterize 1/0 in scientific multiprocessor applications from a variety of production parallel computing platforms and sites. The Charisma project is unique in recording individual read and write requests-in live, multiprogramming, parallel workloads (rather than from selected or nonparallel applications). In this article, we present the first results from the project: a characterization of the file-system workload an iPSC/860 multiprocessor running production, parallel scientific applications at NASA's Ames Research Center.
We present views and analysis of the execution of several PVM codes for Computational Fluid Dynamics on a network of Sparcstations, including (a) NAS Parallel benchmarks CG and MG (White, Alund and Sunderam 1993); (b) a multi-partitioning algorithm for NAS Parallel Benchmark SP (Wijngaart 1993); and (c) an overset grid flowsolver (Smith 1993). These views and analysis were obtained using our Automated Instrumentation and Monitoring System (AIMS) version 3.0, a toolkit for debugging the performance of PVM programs. We will describe the architecture, operation and application of AIMS. The AIMS toolkit contains (a) Xinstrument, which can automatically instrument various computational and communication constructs in message-passing parallel programs; (b) Monitor, a library of run-time trace-collection routines; (c) VK (Visual Kernel), an execution-animation tool with source-code clickback; and (d) Tally, a tool for statistical analysis of execution profiles. Currently, Xinstrument can handle C and Fortran77 programs using PVM 3.2.x; Monitor has been implemented and tested on Sun 4 systems running SunOS 4.1.2; and VK uses X11R5 and Motif 1.2. Data and views obtained using AIMS clearly illustrate several characteristic features of executing parallel programs on networked workstations: (a) the impact of long message latencies; (b) the impact of multiprogramming overheads and associated load imbalance; (c) cache and virtual-memory effects; and (4significant skews between workstation clocks. Interestingly, AIMS can compensate for constant skew (zero drift) by calibrating the skew between a parent and its spawned children. In addition, AIMS' skew-compensation algorithm can adjust timestamps in a way that eliminates physically impossible communications (e.g., messages going backwards in time). Our current efforts are directed toward creating new views to explain the observed performance of PVM programs. Some of the features planned for the near future include: (a) ConfigView, showing the physical topology of the virtual machine, inferred using specially formatted IP (Internet Protocol) packets; and (b) LoadView, synchronous animation of PVM-program execution and resource-utilization patterns.
We present views and analysis of the execution of several PVM (Parallel Virtual Machine) codes for Computational Fluid Dynamics on a networks of Sparcstations, including: (1) NAS Parallel Benchmarks CG and MG; (2) a multi-partitioning algorithm for NAS Parallel Benchmark SP; and (3) an overset grid flowsolver. These views and analysis were obtained using our Automated Instrumentation and Monitoring System (AIMS) version 3.0, a toolkit for debugging the performance of PVM programs. We will describe the architecture, operation and application of AIMS. The AIMS toolkit contains: (1) Xinstrument, which can automatically instrument various computational and communication constructs in message-passing parallel programs; (2) Monitor, a library of runtime trace-collection routines; (3) VK (Visual Kernel), an execution-animation tool with source-code clickback; and (4) Tally, a tool for statistical analysis of execution profiles. Currently, Xinstrument can handle C and Fortran 77 programs using PVM 3.2.x; Monitor has been implemented and tested on Sun 4 systems running SunOS 4.1.2; and VK uses XIIR5 and Motif 1.2. Data and views obtained using AIMS clearly illustrate several characteristic features of executing parallel programs on networked workstations: (1) the impact of long message latencies; (2) the impact of multiprogramming overheads and associated load imbalance; (3) cache and virtual-memory effects; and (4) significant skews between workstation clocks. Interestingly, AIMS can compensate for constant skew (zero drift) by calibrating the skew between a parent and its spawned children. In addition, AIMS' skew-compensation algorithm can adjust timestamps in a way that eliminates physically impossible communications (e.g., messages going backwards in time). Our current efforts are directed toward creating new views to explain the observed performance of PVM programs. Some of the features planned for the near future include: (1) ConfigView, showing the physical topology of the virtual machine, inferred using specially formatted IP (Internet Protocol) packets: and (2) LoadView, synchronous animation of PVM-program execution and resource-utilization patterns.
The Real Time Specification for Java (RTSJ) is an augmentation of Java for real time applications of various degrees of hardness. The central features of RTSJ are real time threads; user defined schedulers; asynchronous events, handlers, and control transfers; a priority inheritance based default scheduler; non-heap memory areas such as immortal and scoped, and non-heap real time threads whose execution is not impeded by garbage collection. The Robust Software Systems group at NASA Ames Research Center has JAVA PATHFINDER (JPF) under development, a Java model checker. JPF at its core is a state exploring JVM which can examine alternative paths in a Java program (e.g., via backtracking) by trying all nondeterministic choices, including thread scheduling order. This paper describes our implementation of an RTSJ profile (subset) in JPF, including requirements, design decisions, and current implementation status. Two examples are analyzed: jobs on a multiprogramming operating system, and a complex resource contention example involving autonomous vehicles crossing an intersection. The utility of JPF in finding logic and timing errors is illustrated, and the remaining challenges in supporting all of RTSJ are assessed.