Due to advances in fiber-optics and VLSI technology, interconnection networks which allow multiple simultaneous broadcasts are becoming feasible. This paper presents the multiprocessor architecture of the Simultaneous...
详细信息
Due to advances in fiber-optics and VLSI technology, interconnection networks which allow multiple simultaneous broadcasts are becoming feasible. This paper presents the multiprocessor architecture of the Simultaneous Optical Multiprocessor Exchange Bus (SOME-Bus), and examines the performance of representative algorithms for matrix operations, merging and sorting. using the message-passing and distributed-shared-memory paradigms. It shows that simple enhancements to the network interface and the cache and directory controllers can result in communication time of 0(l) for the matrix-vector multiplication algorithm using DSM. The SOME-Bus is a low-latency, high-bandwidth, fiber-optic interconnection network which directly links arbitrary pairs of processor nodes without contention, and can efficiently interconnect over 100 nodes. It contains a dedicated channel for the data output of each node, eliminating the need for global arbitration and providing bandwidth that scales directly with the number of nodes in the system. Each of P nodes has an array of receivers, with one receiver dedicated to each node output channel. No node is ever blocked from transmitting by another transmitter or due to contention for shared switching logic. The entire P receiver array can be integrated on a single chip at a comparatively minor cost resulting in O(P) complexity. The SOME-Bus has much more functionality than a crossbar by supporting multiple simultaneous broadcasts of messages, allowing cache consistency protocols to complete much faster. (C) 2003 Elsevier B.V. All rights reserved.
This paper studies the problem of optimal switching surface design for hybrid systems. In particular, a formula is derived for computing the gradient of a given integral performance cost with respect to the switching ...
详细信息
ISBN:
(纸本)0780386825
This paper studies the problem of optimal switching surface design for hybrid systems. In particular, a formula is derived for computing the gradient of a given integral performance cost with respect to the switching surface parameters. The formula reflects the hybrid nature of the system in that it is based on a costate variable having a discrete element and a continuous element. A numerical example with a gradient descent algorithm suggests the potential viability of the formula in optimization.
numerical properties of a computational algorithm for the scalar Bezout identity in the ring of proper stable rational functions is studied in this paper. The method has been implemented in the MATLAB 6.5 computationa...
详细信息
numerical properties of a computational algorithm for the scalar Bezout identity in the ring of proper stable rational functions is studied in this paper. The method has been implemented in the MATLAB 6.5 computational environment using the pre-release version 3.0 of the Polynomnial Toolbox and exposed to extensive numerical testing. The results are summarized and some useful guidelines for using this well known algorithm are given.
This paper studies the problem of optimal switching surface design for hybrid systems. In particular, a formula is derived for computing the gradient of a given integral performance cost with respect to the switching ...
详细信息
This paper studies the problem of optimal switching surface design for hybrid systems. In particular, a formula is derived for computing the gradient of a given integral performance cost with respect to the switching surface parameters. The novelty of the work lies both in the problem formulation itself, and in the introduction of a discrete event variational system for propagating the perturbations across switching surfaces.
Efficient, structure-exploiting techniques for input/output data processing in subspace-based multivariable system identification are investigated. The techniques are implemented in the system identification toolbox f...
详细信息
Efficient, structure-exploiting techniques for input/output data processing in subspace-based multivariable system identification are investigated. The techniques are implemented in the system identification toolbox for discrete-time systems, SLIDENT, incorporated in the freely available Fortran 77 S ubroutine L ibrary i n Co ntrol T heory (SLICOT). Besides drivers and computational routines, this toolbox provides M ATLAB interfaces, implementing several algorithmic approaches. Extensive numerical testing and comparisons with similar MATLAB tools show that SLIDENT is reliable, efficient, and powerful enough to solve industrial identification problems.
Obtention of the phase transmitatnce of the cornea under the paraxial assumption does not provide accurate estimation of the corneal aberrations, for wide apertures of the-pupil. On the other hand, exact ray tracing w...
详细信息
ISBN:
(纸本)0819445967
Obtention of the phase transmitatnce of the cornea under the paraxial assumption does not provide accurate estimation of the corneal aberrations, for wide apertures of the-pupil. On the other hand, exact ray tracing will require a huge amount of calculations band does not provide uniform sampling of the energy distribution at the output. We propose to use a simple approximation that takes into account non-paraxial effects. This approximation allow obtention of a well sampled output plane at no addicional computational cost.
In this paper, we obtain a canonical form for single input n-dimensional linear systems over Z and, in general, over a Bezout domain. We also give an algorithm to obtain that canonical form. (C) 2002 Elsevier Science ...
详细信息
In this paper, we obtain a canonical form for single input n-dimensional linear systems over Z and, in general, over a Bezout domain. We also give an algorithm to obtain that canonical form. (C) 2002 Elsevier Science B.V. All rights reserved.
Calculation of transient temperature distribution in construction elements is the latest topic of interest to engineers today. In many cases, thermal boundary conditions can be applied to only a section of a solid'...
详细信息
Calculation of transient temperature distribution in construction elements is the latest topic of interest to engineers today. In many cases, thermal boundary conditions can be applied to only a section of a solid's boundary, while heat transfer coefficients on boundaries exposed to a moving fluid and fluid temperature can be partially or entirely unknown. A method for solving the aforementioned problems is formulated below;its main advantage is that it does not require any knowledge of, or solution to, the fluid field. Furthermore, it works fast enough to be applied in practice in real time. Additionally, temperature measurement points can be positioned within an element or on the element surface and do not have to form a closed surface. Two examples are included to demonstrate the capabilities of the new approach.
作者：
Paielli, RANASA
Ames Res Ctr Aviat Syst Div Moffett Field CA 94035 USA
Much of the previous literature on conflict resolution is based on instantaneous maneuver models, in which speed and/or heading change dynamics are unmodeled. The effects of the actual maneuver dynamics on the resulti...
详细信息
Much of the previous literature on conflict resolution is based on instantaneous maneuver models, in which speed and/or heading change dynamics are unmodeled. The effects of the actual maneuver dynamics on the resulting minimum separation are analyzed, and a simple numerical algorithm is presented to compensate for those effects. The focus is on level flight in the horizontal plane. Speed changes are modeled as periods of constant along-track acceleration or deceleration, and heading changes are modeled as steady turns of constant rate and radius. These simple kinematic (constrained point-mass) models improve on the resolution accuracy that results from modeling speed and heading changes as instantaneous, but they yield much simpler solutions than general point-mass dynamic models. The accuracy improvement is minor for most heading-change maneuvers, but it is substantial for most speed-change maneuvers. An important operational benefit of the algorithm is that it detects immediately if a conflict is too close to be resolved by a particular maneuver. A method is also outlined for determining the optimal combination of speed and heading change to resolve conflicts. With minor adaptation, the algorithms can also make use of an existing conflict probability estimation algorithm to determine maneuvers for strategic conflict probability reduction.
From a point of view to utilize fracture mechanics of concrete for the solution of engineering problems, the problems in crack modeling, computational algorithm, and damping implementation are discussed in conjunction...
详细信息
From a point of view to utilize fracture mechanics of concrete for the solution of engineering problems, the problems in crack modeling, computational algorithm, and damping implementation are discussed in conjunction with safety assessment of concrete dam against large earthquakes. It is shown that the formulation for the crack-embedded element has an analogy with that of computational plasticity. This analogy enables us to utilize the return-mapping algorithm well established in computational plasticity for the dynamic analysis of crack growth in concrete. The ways of implementing damping to avoid diffused cracking are presented and simple examples of numerical analysis are shown to demonstrate the effect of damping and the performance of appropriate damping implementation for cracked elements. (C) 2002 Elsevier Science Ltd. All rights reserved.
暂无评论