A clustering approach for identifying and quantifying irregularities in interconnection networks

Support for arbitrary topologies has become more popular for system-area networks but very little has been done in trying to characterize their behavior and performance. Traditional parameters like diameter and bisection width are not sufficient for characterizing the irregularities that abound in such networks and fail to give much insight into throughput performance. A clustering approach for partitioning a network into clusters of richly-connected regions is proposed as a means of defining two performance-correlated characterization metrics: intercluster bandwidth index and intercluster link-cost index. The two characterization metrics are shown to have a strong correlation to saturation throughput when link and load distribution of a network is imbalanced. Simulation results also show that the clustering algorithm can be applied to a variety of network configurations and traffic scenarios, particularly irregular ones. With the proposed characterization metrics that correlate more strongly with performance, it is possible to classify networks into categories having similar performance.     

From visionary ideas to products

This paper presents the rationale for the creation of the Microelectronics and Computer Research Corporation (MCC), to be a cooperative research organization funded by a number of U.S. computer and electronics companies.

Factors inhibiting investment in fundamental research by individual companies today are presented and analyzed, and the benefits of the consortium approach for cooperatively performing basic research and early exploratory development work are described.

The present research activities of MCC in the areas of Advanced Computer Architectures, Computer Aided Design, Software Technology, and Packaging and Interconnect technology are presented. This is followed by a discussion of future directions for research in the field of computer science.     

WARRP Core: Optoelectronic Implementation of Network-Router Deadlock-Handling Mechanisms

The wormhole adaptive recovery-based routing via pre-emption(WARRP) core optoelectronic chip, which integrates coredeadlock-handling circuitry for a fully adaptive deadlock-freemultiprocessor network router, is presented. This chip demonstratesprimarily the integration of complex deadlock-recovery circuitry andfree-space optoelectronic input-output on a monolithicGaAs-based chip. The design and implementation of thefirst-generation, bit-serial, torus-connected chip that uses 1400transistors and six light-emitting diode-photodetector pairs is presented.     

A progressive approach to handling message-dependent deadlock in parallel computer systems

Handling deadlocks is essential for providing reliable communication paths between processing nodes in parallel computer systems. The existence of multiple message types and associated inter-message dependencies may cause message-dependent deadlocks in networks that are designed to be free of routing deadlock. Most methods currently used for dealing with message-dependent deadlocks require more system resources than are necessary and/or do not use system resources efficiently. This may have an adverse effect on system performance if resources are scarce. In this paper, we characterize the frequency of message-dependent deadlocks in multiprocessor/multicomputer systems. We also propose a handling technique for message-dependent deadlocks based on progressive deadlock recovery and evaluate its performance with other approaches. Results show that message-dependent deadlocks occur very infrequently under typical circumstances thus, rendering approaches based on avoiding them overly restrictive in the common case. The proposed technique relaxes restrictions considerably, allowing the routing of packets and the handling of message-dependent deadlocks to be much more efficient-particularly when network resources are scarce.     

Comparison of packed-column supercritical fluid chromatography--tandem mass spectrometry with liquid chromatography--tandem mass spectrometry for bioanalytical determination of (R)- and (S)-ketoprofen in human plasma following automated 96-well solid-phase extraction

The popularity of packed-column supercritical fluid, subcritical fluid, and enhanced fluidity liquid chromatographies (pcSFC) for enantiomeric separations has increased steadily over the past few years. The addition of a significant amount (typically 20-95%) of a viscosity lowering agent, such as carbon dioxide, to the mobile phase provides a number of advantages for chiral separations. For example, higher mobile-phase flow rates can often be attained without a concomitant loss in chromatographic efficiency since diffusion coefficients, and optimum velocities, are typically higher in pcSFC. Ultratrace enantioselective quantitation of drugs in biomatrixes is an ideal application for these chromatographic attributes. To demonstrate the utility of this approach, a pcSFC tandem mass spectrometry (pcSFC-MS/MS) method was compared to a LC-MS/MS method for quantitation of the (R)- and (S)-enantiomers of ketoprofen (kt), a potent nonsteroidal, anti-inflammatory drug, in human plasma. After preparation using automated solid-phase extraction in the 96-well format, kt enantiomers were separated on a Chirex 3005 analytical column using isocratic conditions. Validation data and study sample data from patients dosed with either orally or topically administered ketoprofen were generated using both pcSFC and LC as the chromatographic methods to compare and contrast these analytical approaches. Generally, most analytical attributes, including specificity, linearity, sensitivity, accuracy, precision, and ruggedness, for both of these methods were comparable with the exception that the pcSFC separation provided a roughly 3-fold reduction in analysis time. A 2.3-min pcSFC separation and a 6.5-min LC separation provided equivalent, near-baseline-resolved peaks, demonstrating a significant time savings for analysis of large batch pharmacokinetic samples using pcSFC.     

Radiation therapy among atomic bomb survivors, Hiroshima and Nagasaki

Occupational asthma is the most common occupational lung disease in the world. Occupational asthma comprises approximately 25-50% of occupational lung diseases and is responsible for up to 15% of all asthma cases. Either immunologic or nonimmunologic causes may be responsible for occupational asthma. There are approximately 250 compounds known to cause occupational asthma. Common examples include flour, animal dander, isocyanates, and latex. The diagnosis of occupational asthma depends on an accurate history of asthma and documentation that the asthma is caused by workplace exposure. Peak flow measurements are commonly used to provide data to define this relationship. Spirometry and bronchial provocation testing are also helpful. The key management tool in occupational asthma is avoidance of the causative agent. Avoidance is more important than treatment with medications. Occupational asthma can have major socioeconomic impacts on an individual, and the diagnostic work-up and management needs to be performed with this in mind.     

Examination of reinforcement magnitude on the pharmacological disruption of fixed-ratio performance.

Behavioral momentum theory proposes that operant behavior is the product of two separable processes: its rate of occurrence and its resistance to change. Generally speaking, operant situations providing more densely spaced or greater magnitudes of reinforcement should be more resistant to disruption. Attempts to disrupt ongoing behavior by manipulating the availability of food or deprivation level typically have supported the predictions of behavioral momentum. Tests with pharmacological disruptors, however, have yielded mixed results. Most investigations of pharmacological disruption of operant behavior have evaluated momentum across situations that differ in rate of reinforcement. The present experiment was an attempt to systematically replicate prior work, but under conditions of differing reinforcement magnitudes. Pigeons were trained to key peck on a multiple fixed-ratio 30 schedule of food presentation, where different components programmed 2-, 4-, or 8-s access to grain. Resistance to rate-decreasing effects of drugs was evaluated with several compounds drawn from distinct pharmacological classes: chlordiazepoxide, cocaine, clonidine, haloperidol, morphine, and ethanol were tested. Additionally, disruption by prefeeding and extinction was examined. Generally, resistance to change by drug administration was not modulated by reinforcement magnitude. Prefeeding and extinction tests, however, replicated previous work, indicating that our procedure was sensitive to more common disruptors. The results give additional support to the notion that pharmacological disruptors may not behave in the manner predicted by behavioral momentum theory. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Efficient implementation of globally-aware network flow control

Network flow control mechanisms that are aware of global conditions potentially can achieve higher performance than flow control mechanisms that are only locally aware. Owing to high implementation overhead, globally-aware flow control mechanisms in their purest form are seldom adopted in practice, leading to less efficient simplified implementations. In this paper, we propose an efficient implementation of a globally-aware flow control mechanism, called Critical Bubble Scheme, for k-ary n-cube networks. This scheme achieves near-optimal performance with the same minimal buffer requirements of globally-aware flow control and can be further generalized to implement the general class of buffer occupancy-based network flow control. We prove deadlock freedom of the proposed scheme and exploit its use in handling protocol-induced deadlocks in on-chip environments. We evaluate the proposed scheme using both synthetic traffic and real application loads. Simulation results show that the proposed scheme can reduce the buffer access component of packet latency by as much as 62% over locally-aware flow control, and improve average packet latency by 18.8% and overall execution time by 7.2% in full system simulation.     

A formal model of message blocking and deadlock resolution ininterconnection networks

This paper presents a theoretical model of resource allocations and dependencies in wormhole and virtual cut-through interconnection networks. This model allows various types of message blocking to be described precisely, including deadlock. The model distinguishes between messages involved in deadlock and those simply dependent upon deadlock, thus establishing a framework for evaluating the accuracy and correctness of deadlock detection mechanisms. The paper also identifies the necessary and sufficient conditions for the occurrence and resolution of deadlock in interconnection networks, thus providing efficiency and correctness criteria for deadlock resolution mechanisms. Theorems derived from the model are related to various routing algorithms which are based on deadlock recovery