Thomas D. Wagner

The KSR1: experimentation and modeling of poststore

Kendall Square Research introduced the KSR1 system in 1991. The architecture is based on a ring of rings of 64-bit microprocessora. It is a distributed, shared memory system and is scalable. The memory structure is unique and is the key to understanding the system. Different levels of caching eliminates physical memory addressing and leads to the ALLCACHE™ scheme. Since requested data may be found in any of several caches, the initial access time is variable. Once pulled into the local (sub) cache, subsequent access times are fixed and minimal. Thus, the KSR1 is a Cache-Only Memory Architecture (COMA) system.This paper describes experimentation and an analytic model of the KSR1. The focus is on the poststore programmer option. With the poststore option, the programm er can elect to broadcast the updated value of a variable to all processors that might have a copy. This may save time for threads on other processors, but delays the broadcasting thread and places additional traffic on the ring. The specific issue addressed is to determine under what conditions poststore is beneficial. The analytic model and the experimental observations are in good agreement. They indicate that the decision to use poststore depends both on the application and the current system load.

A practical approach to reinforcing concepts in introductory operating systems

The challenge of including both theory and practical experience to reinforce the concepts of theory in Introductory Operating Systems is discussed. The traditional approaches to answering this challenge are presented. A practical approach that uses straightforward programming problems rather than problems that involve modification of an existing operating system is described. Examples of the types of problems used are presented and experience with the proposed approach is described.

Design and analysis of a diffractive optical filter for use in an optoelectronic error-diffusion neural network.

The design, fabrication, experimental characterization, and system-performance analysis of a diffractive optical implementation of an error-diffusion filter for use in digital image halftoning is reported. A diffractive optical filter was fabricated as an eight-level phase element that diffuses the quantization error nonuniformly in both the weighting and the spatial dimensions, according to a prescribed algorithm. Ten identical diffractive elements were fabricated on ten different wafers and subsequently characterized experimentally. A detailed error analysis including both fabrication and instrumentation errors was carried out to quantify the performance of the fabrication process as well as the expected system performance of the filters. Halftone system performance was evaluated by use of the experimental filters performance and both quantitative and qualitative performance metrics. The results of this analysis demonstrate that multiple identical copies of a diffractive optical filter can be produced with sufficient accuracy that no loss in the halftoning system performance results.

The undergraduate capstone software design experience

Ada 95 is being used as the implementation language for a senior level compiler design course at the United States Military Academy. This paper describes experiences and Iessons learned as well as the scenario based approach for a large object oriented compiler project in that course. The term-long compiler project gives the students experience as a part of a (fictitious) designimplementation team building a production compiler. Ada 95 is compared to Pascal and C-H for the same course and positive and negative aspects of the three languages are described.

A highly-parallel mismatch tolerant photonic A/D converter

Summary form only given. Interleaving is a common approach applied to high-speed photonic A/D conversion which reduces the wide-bandwidth input signal to one which can be converted using conventional highspeed A/D converters. The high-speed sampled input is interleaved to N individual channels with each channel operating at 1/N of the sampling rate. These channelization techniques are known to suffer from performance degradations due to channel-to-channel mismatch. Oversampling A/D converters based on /spl utri//spl Sigma/ or error diffusion modulators incorporate low-resolution A/D converters within a feedback architecture to reduce the quantization noise within the signal baseband through spectral noise shaping. We present a typical power spectrum of the output data sequence for a 2nd-order /spl utri//spl Sigma/-modulator.

A partitioning scheme for optoelectronic neural networks

Smart pixel technology provides a promising technological alternative for the implementation of the error diffusion network (EDN) because optical input, electronic processing, and optical output are integrated in a single array. While current smart pixel technology could support 256/spl times/256 array sizes, it is of interest to investigate partitioning approaches which use smaller physical array sizes to achieve the same functionality and performance as larger arrays. One approach to this partitioning is to divide a large image into smaller sub-images, multiplex these sub-images into a small smart pixel EDN, and then demultiplex the partitions into the resulting full-sized image. This concept is demonstrated.

A learning approach to processor allocation in parallel systems

Given a typical parallel system and a collection of applications that are to execute on the system, a common problem is determining an effective allocation of processors among the applications. In this paper a learning approach is applied to processor allocation. The approach is to use a stochastic learning automaton (SLA) as a decision tool. An SLA uses values of the current state description, makes an allocation decision, evaluates its decision at some later time, modifies its decision making process, and tries to find the best allocation strategy by learning from its previous mistakes. The method is applied to the problem of allocating processors to parallel applications in a distributed system such as a cluster of workstations, and is validated through simulation. The result of this study show that a learning approach that utilizes a stochastic learning automaton is effective at making processor allocation decisions in a parallel system.

Experimental characterization of a diffractive optical filter for use in an optoelectronic analog-to-digital converter

The first experimental characterization of an eight-level, phase-only diffractive optical implementation of an error diffusion filter for use in digital image halftoning is reported. Ten identical elements were fabricated through a DARPA-sponsored workshop and subsequently experimentally characterized. A detailed statistical error analysis including both fabrication and instrumentation errors was performed to quantify the performance of the fabrication process as well as the expected system performance of the filters. Over the 10-sample ensemble, an average diffraction efficiency of 83.7% was measured with an RMS error between the designed and experimental coefficients of 0.2411%, producing nearly diffraction limited performance.