The telegram problem was initially proposed as a programming exercise suitable for beginning students. A number of solutions have been published, all containing errors. In this paper the trouble is traced to the original specification and offer a new version. The most significant difference is the use of a grammatical notation for some parts of the specification. The new specification is then followed to a new solution. The concusion is that the new specification is of higher quality, but not...
Topics: DTIC Archive, McKeeman,W M, CALIFORNIA UNIV SANTA CRUZ, *COMPUTER PROGRAMS, *COMPUTER PROGRAMMING,...
This report develops and analyzes a general stochastic model of a production system. The model is closely related to Harrison's (5) assembly-like queueing network, the principal difference being that here we assume all storage buffers have finite capacity. Our attention is focused on a vector stochastic process Z whose components are the contents of the various storage buffers (as functions of time). The principal result is a weak convergence theorem of the type developed by Iglehart and Whitt...
Topics: DTIC Archive, Wenocur,Michael Louis, STANFORD UNIV CA, *QUEUEING THEORY, *PRODUCTION, *MODELS,...
Subject to the accuracy of the acoustic analyzer and the accuracy and completeness of the English Parser, a real-time general solution to the application of English syntactic constraints to spoken English recognition has been developed. This solution is functionally equivalent, in many ways, to the syntax processing of spoken English in the human brain. Because it closely models the syntax processing of the Human Speech Recognition System (HSRS), it is most effective when used with the several...
Topics: DTIC Archive, Routh,R L, AIR FORCE INST OF TECH WRIGHT-PATTERSON AFB OH SCHOOL OF ENGINEERING,...
Recent trends have exposed three key problems in today's operating systems. The first is the emergence of I/O latency as the dominant factor in the performance of many applications. The second is the need to cope with mobile communication environments where bandwidth and latency may be highly variable. The third is the importance of search activity to locating files of interest in a distributed system. In this paper we describe a single unifying abstraction called dynamic sets which can offer...
Topics: DTIC Archive, Steere, David, CARNEGIE-MELLON UNIV PITTSBURGH PA DEPT OF COMPUTER SCIENCE,...
The trend towards distributed Command and Control systems in Naval platforms necessitates the provision of Local Area Network performance measurement tools and techniques. A special purpose network analyser has been developed within ITD to measure the performance of a Fiber Distributed Data Interface (FDDI) network. This paper describes the architecture and operation of the network analyser. This work has been conducted to support the Directorate of Naval Combat Systems Engineering in their...
Topics: DTIC Archive, Allwright, Alan, DEFENCE SCIENCE AND TECHNOLOGY ORGANIZATION CANBERRA (AUSTRALIA),...
This final report describes research in optical neural networks performed at Hughes Research Laboratories under a three year DARPA sponsored contract the advantages of optics for neural network implementations, including high storage capacity, connectivity, and very fine grained parallelism, was demonstrated. The optical neurocomputer developed under this program is based on a new type of holography which we call multiple grating holography, in which this approach reduces crosstalk and improves...
Topics: DTIC Archive, Owechko, Yuri, HUGHES RESEARCH LABS MALIBU CA, *NEURAL NETS, *HOLOGRAPHY, *ARTIFICIAL...
The message handling systems which have been developed and used within the ARPANET have spurred widespread interest. We are studying extensions of this technology, contemplating a system which will provide a message handling service of large capacity and high reliability while meeting stringent security requirements. Our task is to explore the suitability of the Pluribus computer for this application. The Pluribus multiprocessor has many characteristics which make it attractive as a machine for...
Topics: DTIC Archive, Heart, Frank E, BOLT BERANEK AND NEWMAN INC CAMBRIDGE MA, *COMPUTER COMMUNICATIONS,...
Optical materials for spectral hole burning and coherent transient applications were designed and evaluated, emphasizing relevance to devices including high performance memory, signal processing, high speed holographic recording and analysis, and optical switches, routers, and buffers. The ultimate limits on material performance set by atomic mechanisms were determined. The developed materials contained Eu3+, Pr3+, Tm3+, Tb3+, or Er3+ active ions in various oxide hosts. The first report was...
Topics: DTIC Archive, Cone, Rufus L., MONTANA STATE UNIV BOZEMAN DEPT OF PHYSICS, *OPTICAL MATERIALS,...
Parallelization has become a popular mechanism to speed up data classification tasks that deal with large amounts of data. This paper describes a high level, fine grained parallel formulation of a decision tree based classifier for memory resident datasets on SMPs. We exploit two levels of divide and conquer parallelism in the tree builder: at the outer level across the tree nodes, and at the inner level within each tree node. Lightweight Pthreads are used to express this highly irregular and...
Topics: DTIC Archive, Narlikar, Girija J., CARNEGIE-MELLON UNIV PITTSBURGH PA DEPT OF COMPUTER SCIENCE,...
The Digital Avionics Information System (DAIS) Test Facility requires a means for exercising the serial output and input interface devices of the remote terminal. This report describes the design and operation of the unit needed to perform this function. (Author)
Topics: DTIC Archive, Caffrey,James C, AIR FORCE AVIONICS LAB WRIGHT-PATTERSON AFB OHIO, *INPUT OUTPUT...
The Joint Transform Correlator was built using a combination of optical components, analog electronics, and digital electronics. The optical system breadboard can be used to evaluate the correlation process and determine tolerance limits of the optical process. Correlation is achieved using a reference matched filter adjacent to the input image. (Author)
Topics: DTIC Archive, Brown,Hugh B, AMPEX CORP REDWOOD CITY CA DATA SYSTEMS DIV, *IMAGE PROCESSING,...
The user manual for the 2-D graphics toolkit for a transputer based parallel processor is presented. The toolkit consists of a package of 2-D display routines that can be used for the simulation visualizations. It supports multiple windows, double buffered screens for animations, and simple graphics transformations such as translation, rotation, and scaling. The display routines are written in occam to take advantage of the multiprocessing features available on transputers. The package is...
Topics: NASA Technical Reports Server (NTRS), APPLICATIONS PROGRAMS (COMPUTERS), COMPUTER GRAPHICS,...
The operating instructions for the magnetic bearings of a high-performance optical disk buffer are provided. Among the topics that are discussed are the following: front panel layout, turn-on procedure, shut-down procedure, and latch-up protection. Additionally, comprehensive engineering drawings are presented for the design.
Topics: NASA Technical Reports Server (NTRS), BUFFER STORAGE, ENGINEERING DRAWINGS, EQUIPMENT...
An optical disk buffer concept can provide gigabit-per-second data rates and terabit capacity through the use of arrays of solid state lasers applied to a stack of erasable/reusable optical disks. The RCA optical disk buffer has evoked interest by NASA for space applications. The porous graphite air bearings in the rotary spindle as well as those used in the linear translation of the read/write head would be replaced by magnetic bearings or mechanical (ball or roller) bearings. Based upon past...
Topics: NASA Technical Reports Server (NTRS), ACTIVE CONTROL, BUFFER STORAGE, ELECTRONIC CONTROL,...
A method and apparatus is presented for reformatting and de-serializing a serially-received sequence of data words, each consisting of a fixed number of binary data bits. A block of nm bits is serially fed into a shift register or serially-connected group of shift registers. In lieu of the(nm-1)th shifts, the bits are rearranged within the shift register in parallel fashion, according to a prescribed scheme. Shifting then continues, until the first bit of each data word appears in the last bit...
Topics: NASA Technical Reports Server (NTRS), DIGITAL DATA, FORMAT, PARALLEL PROCESSING (COMPUTERS),...
The software package developed for the KIM-1 Micro-System and the Mini-L PLL receiver to simplify taking flight test data is described along with the address and data bus buffers used in the KIM-1 Micro-system. The interface hardware and timing are also presented to describe completely the software programs.
Topics: NASA Technical Reports Server (NTRS), BUFFER STORAGE, COMPUTER SYSTEMS PROGRAMS, FLIGHT TESTS,...
From the bitsavers.org collection, a scanned-in computer-related document. ibm :: 303x :: GA22-7060-3 3033 FuncChar Jan79
Topics: processor, storage, channel, ibm, data, console, control, buffer, instruction, channels, processor...
From the bitsavers.org collection, a scanned-in computer-related document. ibm :: 1620 :: fe :: 227-5816-0 1620 Model 1 1622 Card Read Punch Feature Customer Engineering Manual of Instruction Sep63
Topics: punch, trigger, latch, turned, buffer, data, punch buffer, card punch, power supply, read buffer,...
From the bitsavers.org collection, a scanned-in computer-related document. univac :: uss :: SolidState90 IOUnits
Topics: instruction, buffer, storage, word, punch, print, interlace, card cycle, main storage, read...
From the bitsavers.org collection, a scanned-in computer-related document. amdahl :: datapro :: 70C-035MM-10 8509 Amdahl 580
Topics: channels, amdahl, storage, bytes, data, memory, main, processor, console, cpu, main memory, main...
Recently, a compiler-assisted approach to multiple instruction retry was developed. In this scheme, a read buffer of size 2N, where N represents the maximum instruction rollback distance, is used to resolve one type of data hazard. This hardware support helps to reduce code growth, compilation time, and some of the performance impacts associated with hazard resolution. The 2N read buffer size requirement of the compiler-assisted approach is worst case, assuring data redundancy for all data...
Topics: NASA Technical Reports Server (NTRS), BUFFER STORAGE, COMPILERS, COMPUTERIZED SIMULATION, FAULT...
A device driver for an EMR 760 BDC was written for the VAX-11/780. An Application program was written to use the driver and transfer rates in excess of 190,000 sixteen bit words per second were realized. The package successfully transferred 10 Mb to disk using DMA operations. The techniques for writing device drivers on the VAX-11/780 are discussed. (Author)
Topics: DTIC Archive, Rall,David L, AIR FORCE INST OF TECH WRIGHT-PATTERSON AFB OH SCHOOL OF ENGINEERING,...
A device driver for an EMR 760 BDC was written for the VAX-11/780. An Application program was written to use the driver and transfer rates in excess of 190,000 sixteen bit words per second were realized. The package successfully transferred 10 Mb to disk using DMA operations. The techniques for writing device drivers on the VAX-11/780 are discussed. (Author)
Topics: DTIC Archive, Rall,David L, AIR FORCE INST OF TECH WRIGHT-PATTERSON AFB OH SCHOOL OF ENGINEERING,...
A device driver for an EMR 760 BDC was written for the VAX-11/780. An Application program was written to use the driver and transfer rates in excess of 190,000 sixteen bit words per second were realized. The package successfully transferred 10 Mb to disk using DMA operations. The techniques for writing device drivers on the VAX-11/780 are discussed. (Author)
Topics: DTIC Archive, Rall,David L, AIR FORCE INST OF TECH WRIGHT-PATTERSON AFB OH SCHOOL OF ENGINEERING,...
Synchronous time division multiplexing is analyzed. Packets of information arrive at the system as a compound Poisson process, and are transmitted only during individual periodic intervals. Packet arrivals are blocked (lost) if the system has a finite capacity and is congested. Using the theory of semiregenerative processes, the distribution of the number of packets in the system (system size) is found. This nonstationary distribution is used to determine the complete system behavior, including...
Topics: NASA Technical Reports Server (NTRS), BUFFER STORAGE, PACKET TRANSMISSION, QUEUEING THEORY, TIME...
Developing memory systems to support high-speed processors is a major challenge to computer architects. Cache memories can improve system performance but the latency of main memory remains a major penalty for a cache-miss. A novel approach to improve system performance is the use of a memory prediction buffer. The memory prediction buffer (MPB) is inserted between the cache and main memory. The MPB predicts the next cache-miss address and pre-fetches the data. The use of an MPB in a computer...
Topics: DTIC Archive, Billingsley, Jr, Arthur B, NAVAL POSTGRADUATE SCHOOL MONTEREY CA, *BUFFERS, *SYSTEMS...
The software required to support a new technique for acquiring large amounts of optical data in a short time period has been developed. The data acquisition system incorporates an image intensifier and vidicon tube in combination. It is controlled by a minicomputer utilizing the software described. The software handles all control of the image-intensifier-vidicon system, acquires the data from the system, and stores the data for subsequent retrieval and reduction. The software incorporates a...
Topics: DTIC Archive, Jones, J H, ARO INC ARNOLD AFS TN, *COMPUTER APPLICATIONS, *IMAGE...
This report describes the feasibility of performing the conversion of data from a field deployable data base to input for a radar scene generation algorithm in the field within a less than 10-minute period. Although the method described is not the only method, it is one which will work within the time allotted and fulfills other criteria, such as amount of disc storage required and computer power available. (Author)
Topics: DTIC Archive, AUTOMETRIC INC FALLS CHURCH VA, *SIGNAL PROCESSING, *RADAR SIGNALS, *TERRAIN ANALYSIS...
This document presents the design changes that allows a PLANIT lesson to control ACC/OCC, VFMED, and MIOD terminals so that TACFIRE/TOS2 systems operations can be simulated under PLANIT.
Topics: DTIC Archive, LITTON SYSTEMS INC VAN NUYS CA DATA SYSTEMS DIV, *COMPUTER AIDED INSTRUCTION,...
Numerical applications frequently contain nested loops that process large arrays of data. The execution of these loop structures often produces memory reference patterns that utilize data caches poorly. Indeed, poor reuse of the data, large working set sizes, and frequent non-unit stride accesses all combine to cause many cache misses. To improve cache performance, data copying has been proposed. However, this technique has high overhead. In this paper, instead, we propose a combined hardware...
Topics: DTIC Archive, Yamada, Yoji, ILLINOIS UNIV AT URBANA COORDINATED SCIENCE LAB, *SOFTWARE ENGINEERING,...
This program is an attempt at a large scale, multi-level general circulation model of the atmosphere. The original conception was to implement a very flexible program. The major programs are: INITE (Since a typical problem will require more storage than can be accomodated, this program parcels the data and codes information as to how it is partitioned.); GEX (A solution program to step the fields ahead in time.); and DISPLY (This program displays the various fields which have been saved from...
Topics: DTIC Archive, Leonard,Thomas J, REGIS COLL WESTON MASS, *COMPUTER PROGRAMS, *ATMOSPHERIC MOTION,...
A device driver for an EMR 760 BDC was written for the VAX-11/780. An Application program was written to use the driver and transfer rates in excess of 190,000 sixteen bit words per second were realized. The package successfully transferred 10 Mb to disk using DMA operations. The techniques for writing device drivers on the VAX-11/780 are discussed. (Author)
Topics: DTIC Archive, Rall,David L, AIR FORCE INST OF TECH WRIGHT-PATTERSON AFB OH SCHOOL OF ENGINEERING,...
The message handling systems which have been developed and used within the ARPANET have spurred widespread interest. We are studying extensions of this technology, contemplating a system which will provide a message handling service of large capacity and high reliability while meeting stringent security requirements. Our task is to explore the suitability of the Pluribus computer for this application. The Pluribus multiprocessor has many characteristics which make it attractive as a machine for...
Topics: DTIC Archive, Heart, Frank E, BOLT BERANEK AND NEWMAN INC CAMBRIDGE MA, *COMPUTER COMMUNICATIONS,...
Topics: DTIC Archive, COMPUTER CORP OF AMERICA CAMBRIDGE MA, *DATA STORAGE SYSTEMS, *INFORMATION RETRIEVAL,...
The performance of demand-driven caching is known to depend on the locality of reference exhibited by the stream of requests made to the cache. In spite of numerous efforts, no consensus has been reached on how to formalize this notion, let alone on how to compare streams of requests on the basis of their locality of reference. The authors take on this issue with an eye towards validating operational expectations associated with the notion of locality of reference. They focus on two folk...
Topics: DTIC Archive, Makowski, Armand M, MARYLAND UNIV COLLEGE PARK INST FOR SYSTEMS RESEARCH,...
Recently, distributed files systems have aggressively exploited caching to provide high availability to their clients. Such systems allow users to access and modify cached data even when clients become disconnected (e.g. a laptop removed from its docking station) or weakly connected (e.g. a laptop connected via a modem) to servers. In doing so, these file systems violate one of the key properties of caching: transparency. When disconnected or weakly connected, these systems cannot always...
Topics: DTIC Archive, Ebling, Maria R., CARNEGIE-MELLON UNIV PITTSBURGH PA SCHOOL OF COMPUTER SCIENCE,...
The message handling systems which have been developed and used within the ARPANET have spurred widespread interest. We are studying extensions of this technology, contemplating a system which will provide a message handling service of large capacity and high reliability while meeting stringent security requirements. Our task is to explore the suitability of the Pluribus computer for this application. The Pluribus multiprocessor has many characteristics which make it attractive as a machine for...
Topics: DTIC Archive, Heart, Frank E, BOLT BERANEK AND NEWMAN INC CAMBRIDGE MA, *COMPUTER COMMUNICATIONS,...
In this report, we summarize the results of our small buffer project. The goals of the project were (1) to model the behavior of TCP in a network where the routers have very small buffers, (2) to determine a rule for sizing buffers in such networks, and (3) to improve on TCP so that it can operate well in such a network. The main outcomes of the project were as follows: We developed an algorithm based on the use of rational approximations coupled with a Hierarchical Markovian model of network...
Topics: DTIC Archive, Towsley, Don, MASSACHUSETTS UNIV AMHERST DEPT OF COMPUTER SCIENCE, *INTERNET, *BUFFER...
Resource allocation is an issue in any system for which resources are in short supply. In computer networks, the resources in question are bandwidth, buffer space, and processing time, and they are allocated at multiple scales of time and quantity (e.g. from a single user's packets over a 30 second interval to a network providers backbone capacity for a 1 year interval). For military networks and computers, these resources may be scarce or limited during attack, at remote sites, etc. Decisions...
Topics: DTIC Archive, Reeves, Douglas S, NORTH CAROLINA STATE UNIV AT RALEIGH DEPT OF COMPUTER SCIENCE,...
The drive towards richer and more interactive web content places increasingly stringent requirements on datacenter networks. The speed with which such networks respond to packet drops limits their ability to meet high-percentile flow completion time SLOs. Indirect notifications to packet drops (e.g., duplicates in an end-to-end acknowledgment sequence) are an important limitation to the agility of response to packet drops. We propose FastLane, a new in-network drop notification mechanism....
Topics: DTIC Archive, CALIFORNIA UNIV BERKELEY DEPT OF ELECTRICAL ENGINEERING AND COMPUTER SCIENCES,...
The paper gives an overview of commonly used synchronization primitives and literature, and presents a new form of primitive expressing conditional critical regions. A new solution is presented to the problem of readers and writers, utilizing the proposed synchronization primitive. The solution is simpler and shorter than other known algorithms. The first sections of the paper give a tutorial introduction into established methods, in order to provide a suitable background for the remaining...
Topics: DTIC Archive, Pettersen, Odd, STANFORD UNIV CA DEPT OF COMPUTER SCIENCE, *ARTIFICIAL INTELLIGENCE,...
An adaptive transform coding algorithm based on a recursive procedure in the transform domain has been developed. Both the quantization parameters and bit assignment are dynamically determined and thus are closely matched to the actual image structure. Overhead requirements are minimal.
Topics: DTIC Archive, Tescher, Andrew G, AEROSPACE CORP EL SEGUNDO CA ENGINEERING SCIENCE OPERATIONS,...
Three Input Buffer designs which provide intermediate storage between the Analog to Digital Converters and the Digital Matched Filter of a Radar Signal Processor are presented. All desings use a basic all-ECL buffer module. Prototype hardware experiments indicate that with 8:1 input data multiplexing, ECL 10K technology will yield input rates up to 142 Ms/s per channel, and output rates of 45 Ms/s per convolver rail. (Author)
Topics: DTIC Archive, Huntoon,Albert H, MASSACHUSETTS INST OF TECH LEXINGTON LINCOLN LAB, *SIGNAL...
A pipeline as defined here is a collection of segments of hardware which can operate simultaneously. A task flows synchronously from segment to segment for its execution. Each task follows one of several distinct task flow patterns which are assumed to be fixed and known in advance. It is characteristic of pipelines that a task can be initiated in the pipeline before an earlier initiated task has completed its execution. A problem arises when two or more tasks try to use the same segment at the...
Topics: DTIC Archive, Patel,Janek H, ILLINOIS UNIV AT URBANA-CHAMPAIGN COORDINATED SCIENCE LAB, *PARALLEL...
This report covers work performed during the first year of the extension of the ARPANET Routing Algorithm Improvements Contract. Network buffer management issues are discussed and a new buffer management scheme for the ARPANET is designed. Logical addressing is discussed, and a design is given for a logical addressing scheme suitable for ARPANET or DIN II. The applicability of ARPANET Routing to DIN II is evaluated. The possibility of extending ARPANET's routing algorithm to provide multiple...
Topics: DTIC Archive, Rosen, E. C., BOLT BERANEK AND NEWMAN INC CAMBRIDGE MA, *COMPUTER COMMUNICATIONS,...
This thesis applies the state of the art techniques for methodical design of secure operating systems to a distributed, multi-microprocessor environment. Explicit process structure and utilization of virtual environments are the fundamental concepts that form a basis for the design presented. The primary design techniques utilized in the design are segmentation, distributed operating system, security kernel, multiprocessing, 'cache' memory strategy and multiprogramming. The resulting design is...
Topics: DTIC Archive, O'Connell,James Steven, NAVAL POSTGRADUATE SCHOOL MONTEREY CA, *MICROPROCESSORS,...
A high-speed, special-purpose preprocessor has been designed for the GEODSS (Ground-based Electro-Optical Deep Space Surveillance) system during the experimental test and evaluation phase of this program. The preprocessor is part of snapshot MTI technique presently being tested and evaluated at the GEODSS experimental Test Site (ETS). The preprocessor provides the interface between a low light level TV sensor and the MITRE multi-minicomputer processor. The primary functions performed by the...
Topics: DTIC Archive, Bungard, D. R., MITRE CORP BEDFORD MA, *SPACE SURVEILLANCE, *PREPROCESSING, *MOVING...
A programmable symbol generator, in-raster buffer refresh memory and software development station was designed, fabricated and tested. The system was developed to meet the Navy's needs for improved reliability vertical situation displays for advanced weapon systems. This report documents the last phase of a four phase study and development program sponsored by NADC. Also included in this report is a discussion of color requirements and a survey of color display technology. (Author)
Topics: DTIC Archive, Wolfson,G, HUGHES AIRCRAFT CO CULVER CITY CALIF DISPLAY SYSTEMS LAB, *DISPLAY...
148
148
Dec 15, 2009
12/09
by
National Semiconductor Corporation
texts
eye 148
favorite 1
comment 0
"400029-2"--Cover
Topics: Integrated circuits, Interface circuits, Data transmission systems, INTEGRATED CIRCUITS,...
The objectives of this research were the investigation of certain issues on routing, flow control, and the relationship between the two. More specifically, the following research tasks were carried out: (1) It was shown that while deterministic routing may seem to perform better under stable traffic conditions, the adaptive strategy out-performs the deterministic one for networks spanning several time zones; (2) Approximation models were determined for determining response time on a virtual...
Topics: DTIC Archive, Chou,Wushow, NORTH CAROLINA STATE UNIV AT RALEIGH, *COMPUTER COMMUNICATIONS,...
