Structuring fault-tolerant object systems for modularity in adistributed environment

The object-oriented approach to system structuring has found widespread acceptance among designers and developers of robust computing systems. The authors propose a system structure for distributed programming systems that support persistent objects and describe how properties such as persistence and recoverability can be implemented. The proposed structure is modular, permitting easy exploitation of any distributed computing facilities provided by the underlying system. An existing system constructed according to the principles espoused here is examined to illustrate the practical utility of the proposed approach to system structuring     

On the performance of synchronized programs in distributed networkswith random processing times and transmission delays

A synchronizer is a compiler that transforms a program designed to run in a synchronous network into a program that runs in an asynchronous network. The behavior of a simple synchronizer, which also represents a basic mechanism for distributed computing and for the analysis of marked graphs, was studied by S. Even and S. Rajsbaum (1990) under the assumption that message transmission delays and processing times are constant. We study the behavior of the simple synchronizer when processing times and transmission delays are random. The main performance measure is the rate of a network, i.e., the average number of computational steps executed by a processor in the network per unit time. We analyze the effect of the topology and the probability distributions of the random variables on the behavior of the network. For random variables with exponential distribution, we provide tight (i.e., attainable) bounds and study the effect of a bottleneck processor on the rate     

E-kernel: an embedding kernel on the IBM victor V256 multiprocessorfor program mapping and network reconfiguration

We present the design of E-kernel, an embedding kernel on the Victor V256 message-passing partitionable multiprocessor, developed for the support of program mapping and network reconfiguration. E-kernel supports the embedding of a new network topology onto Victor's 2D mesh and also the embedding of a task graph onto the 2D mesh network or the reconfigured network. In the current implementation, the reconfigured network can be a line or an even-size ring, and the task graphs meshes or tori of a variety of dimensions and shapes or graphs with similar topologies. For application programs having these task graph topologies and that are designed according to the communication model of E-kernel, they can be run without any change on partitions connected by the 2D mesh, line, or ring. Further, E-kernel attempts the communication optimization of these programs on the different networks automatically, thus making both the network topology and the communication optimization attempt completely transparent to the application programs. Many of the embeddings used in E-kernel are optimal or asymptotically optimal (with respect to minimum dilation cost). The implementation of E-kernel translated some of the many theoretical results in graph embeddings into practical tools for program mapping and network reconfiguration in a parallel system. E-kernel is functional on Victor V256. Measurements of E-kernel's performance on V256 are also included     

Using recurrent neural networks for adaptive communication channelequalization

Nonlinear adaptive filters based on a variety of neural network models have been used successfully for system identification and noise-cancellation in a wide class of applications. An important problem in data communications is that of channel equalization, i.e., the removal of interferences introduced by linear or nonlinear message corrupting mechanisms, so that the originally transmitted symbols can be recovered correctly at the receiver. In this paper we introduce an adaptive recurrent neural network (RNN) based equalizer whose small size and high performance makes it suitable for high-speed channel equalization. We propose RNN based structures for both trained adaptation and blind equalization, and we evaluate their performance via extensive simulations for a variety of signal modulations and communication channel models. It is shown that the RNN equalizers have comparable performance with traditional linear filter based equalizers when the channel interferences are relatively mild, and that they outperform them by several orders of magnitude when either the channel's transfer function has spectral nulls or severe nonlinear distortion is present. In addition, the small-size RNN equalizers, being essentially generalized IIR filters, are shown to outperform multilayer perceptron equalizers of larger computational complexity in linear and nonlinear channel equalization cases.     

Connectivity and performance tradeoffs in the cascade correlationlearning architecture

The cascade correlation is a very flexible, efficient and fast algorithm for supervised learning. It incrementally builds the network by adding hidden units one at a time, until the desired input/output mapping is achieved. It connects all the previously installed units to the new unit being added. Consequently, each new unit in effect adds a new layer and the fan-in of the hidden and output units keeps on increasing as more units get added. The resulting structure could be hard to implement in VLSI, because the connections are irregular and the fan-in is unbounded. Moreover, the depth or the propagation delay through the resulting network is directly proportional to the number of units and can be excessive. We have modified the algorithm to generate networks with restricted fan-in and small depth (propagation delay) by controlling the connectivity. Our results reveal that there is a tradeoff between connectivity and other performance attributes like depth, total number of independent parameters, and learning time.     

Fixed-order H∞ compensator design

A method is presented for the construction of fixed-order compensators to provide H_∞ norm constraint for linear control systems with exogenous disturbances. The method is based on the celebrated bounded-real lemma that predicates the H_∞ norm constraint via a Riccati inequality. The synthesis of fixed-order controllers whose dimensions are less than the order of a given plant, is demonstrated by a set of sufficient conditions along with a numerical algorithm.     

Low temperature wafer direct bonding

A pronounced increase of interface energy of room temperature bonded hydrophilic Si/Si, Si/SiO₂, and SiO₂/SiO ₂ wafers after storage in air at room temperature, 150°C for 10-400 h has been observed. The increased number of OH groups due to a reaction between water and the strained oxide and/or silicon at the interface at temperatures below 110°C and the formation of stronger siloxane bonds above 110°C appear to be the main mechanisms responsible for the increase in the interface energy. After prolonged storage, interface bubbles are detectable by an infrared camera at the Si/Si bonding seam. Desorbed hydrocarbons as well as hydrogen generated by a reaction of water with silicon appear to be the major contents in the bubbles. Design guidelines for low temperature wafer direct bonding technology are proposed     

Flexible, dry-released process for aluminum electrostatic actuators

We present an all-aluminum MEMS process (Al-MEMS) for the fabrication of large-gap electrostatic actuators with process steps that are compatible with the future use of underlying, pre-fabricated CMOS control circuitry. The process is purely additive above the substrate as opposed to processes that depend on etching pits into the silicon, and thereby permits a high degree of design freedom. Multilayer aluminum metallization is used with organic sacrificial layers to build up the actuator structures. Oxygen-based dry etching is used to remove the sacrificial layers. While this approach has been previously used by other investigators to fabricate optical modulators and displays, the specific process presented herein has been optimized for driving mechanical actuators with relatively large travels. The process is also intended to provide flexibility for design and future enhancements. For example, the gap height between the actuator and the underlying electrode(s) can be set using an adjustable polyimide sacrificial layer and aluminum “post” deposition step. Several Al-MEMS electrostatic structures designed for use as mechanical actuators are presented as well as some measured actuation characteristics     

MEETING THE IT CHALLENGES OF BUSINESS The Changing Mandate for MBAs

What is the implication for business when information technology (IT) changes in the workplace without a commensurate change in the composition of business programs educating tomorrow's employees? A survey of MBA graduates forms the basis of this article on the IT skills needed in the marketplace.     

Generalized chebyshev inequality and its application to survivability analysis of complex systems

Conclusions The generalized Chebyshev inequalities are of independent value in mathematical analysis, probability theory, and other fields. Survivability analysis of elements and systems requires specification of functional probability characteristics-distributions of the current durability point. Probabilistic calculation of survivability of complex systems can be carried out using logical-probabilistic methods [22, 28], because the probabilistic-physical meaning of the distribution of current durability point at the point x is the failure probability of an element (a system) given the deterministic level x of the next shock. The methodology of reliability theory can be updated by focusing on a physical stochastic process, instead of time to failure, as the cause of failure. In conclusion, I would like to thank I. A. Ibragimov for discussion of results and some useful comments. Translated from Kibernetika i Sistemnyi Analiz, No. 2, pp. 159–166, March–April, 1994.