Automated dynamic detection of busy–wait synchronizations

With the advent of multicores, multithreaded programming has acquired increased importance. In order to obtain good performance, the synchronization constructs in multithreaded programs need to be carefully implemented. These implementations can be broadly classified into two categories: busy–wait and schedule‐based. For shared memory architectures, busy–wait synchronizations are preferred over schedule‐based synchronizations because they can achieve lower wakeup latency, especially when the expected wait time is much shorter than the scheduling time. While busy–wait synchronizations can improve the performance of multithreaded programs running on multicore machines, they create a challenge in program debugging, especially in detecting and identifying the causes of data races. Although significant research has been done on data race detection, prior works rely on one important assumption—the debuggers are aware of all the synchronization operations performed during a program run. This assumption is a significant limitation as multithreaded programs, including the popular SPLASH‐2 benchmark have busy–wait synchronizations such as barriers and flag synchronizations implemented in the user code. We show that the lack of knowledge of these synchronization operations leads to unnecessary reporting of numerous races. To tackle this problem, we propose a dynamic technique for identifying user‐defined synchronizations that are performed during a program run. Both software and hardware implementations are presented. Furthermore, our technique can be easily exploited by a record/replay system to significantly speedup the replay. It can also be leveraged by a transactional memory system to effectively resolve a livelock situation. Our evaluation confirms that our synchronization detector is highly accurate with no false negatives and very few false positives. We further observe that the knowledge of synchronization operations results in 23% reduction in replay time. Finally, we show that using synchronization knowledge livelocks can be efficiently avoided during runtime monitoring of programs. Copyright © 2009 John Wiley & Sons, Ltd.     

Thermal transients during processing of materials by very high power ultrasonic additive manufacturing

Dynamic recrystallisation at interfaces has been suggested as the bonding mechanism in the joining of metallic tapes, during very high power ultrasonic additive manufacturing. To understand the reasons for such occurrence of dynamic recrystallization, thermal transients from the interface regions were recorded during processing of aluminum alloy (3003 and 6061 series) and 11 000 copper tapes under similar conditions. Measurements in 3003 Al were also carried out for different processing parameters. Measured peak temperatures were seen to increase with increase in shear strength of the material and ultrasonic vibration amplitude. The observations have been rationalized based on interfacial heating at asperities due to adiabatic plastic deformation.     

Global asymptotic stability analysis for neutral-type complex-valued neural networks with random time-varying delays

This paper investigates the global asymptotic stability problem for a class of neutral-type complex-valued neural networks with random time-varying delays. By introducing a stochastic variable with Bernoulli distribution, the information of time-varying delay is assumed to be random time-varying delays. By constructing an appropriate Lyapunov–Krasovskii functional and employing inequality technique, several sufficient conditions are obtained to ensure the global asymptotically stability of equilibrium point for the considered neural networks. The obtained stability criterion is expressed in terms of complex-valued linear matrix inequalities, which can be simply solved by effective YALMIP toolbox in MATLAB. Finally, three numerical examples are given to demonstrate the efficiency of the proposed main results.     

Effect of embedment length of untreated natural fibres on the bond behaviour in cement mortar

The present investigation is focused on studying the effects of various matrices with 1:3, 1:4 and 1:5 mortars and fibre types of sisal and coir on the bond behavior at various ages of curing, i.e., 24 h, 3 d, 7 d and 28 d. The other parameters included in the investigation are water/cement (w/c) ratio, sand gradation and embedment length of fibres. In addition, the type of failure of sisal and coir fibres for different mixes of mortars at various curing ages is also reported. From the results, it is seen that the bond strength is improving with respect to age of curing in case of sisal fibres, but decreases in case of coir fibres. The failure of fibres due to fibre fracture is observed in sisal fibres and fibre pullout is observed in coir fibres. The other varying parameters such as mortar mixes, sand gradation, w/c ratio and embedded length also showed significant effect on bond behaviour of sisal and coir fibre with the cement mortar mixes.     

Influence of ultrasonic cavitation on surface residual stresses in AISI 304 stainless steel

The effect of ultrasonic cavitation in water on residual stress changes in AlSl 304 stainless steel has been investigated. Studies indicate that high-intensity ultrasonic cavitation introduces a very high compressive residual stress at the surface (due to work-hardening) even for short durations of exposure at ambient temperatures. With increased exposure, the stresses become more compressive; however, they tend to reach a saturation value. Different combinations of temperature, time and cavitation intensity were tried out and the best effects were noticed for a treatment temperature of 5 °C. 304 stainless steel was chosen for the present study on account of its amenability to strong work-hardening. The test specimen was attached to the tip of an ultrasonic vibrator and immersed in the cavitating liquid, i.e. water. However, even in situations where the specimen was kept in a stand-off position close to the vibrator tip (with water in between) similar effects were noticed. The maximum depth of hardening was found to be about 70 m. During this process, there was also a mild roughening of the surface. An incidental observation pertains to the formation of both and martensites at the surface detectable by X-ray diffractometer recordings for specific conditions of cavitation treatment. The required high intensities of vibration in this study were obtained through an in-house built highpower ultrasonic generator working at a frequency around 20 kHz.     

Quantitative Evaluation of Bulk and Interface Microstructures in Al-3003 Alloy Builds Made by Very High Power Ultrasonic Additive Manufacturing

Ultrasonically consolidated 3003 aluminum alloy builds were prepared with constituent tapes by using a very high power ultrasonic additive manufacturing (UAM) process. Microstructures of interface and bulk regions were quantitatively characterized using the electron backscattered diffraction technique. The interface microstructure consists of equiaxed grains. The 〈111〉 crystallographic directions of these grains were aligned with the normal direction of the specimen, confirming a shear deformation mode at these regions. In addition, due to recrystallization, the density of low-angle grain boundaries also significantly decreased. In contrast, original elongated grains and partially recrystallized grains were observed in the bulk region of the tape. These elongated grains correspond to rolling texture components of face-centered-cubic materials. The preceding microstructure gradients are rationalized based on the accumulated thermomechanical cycles during processing.     

Robust Passivity and Stability Analysis of Uncertain Complex-Valued Impulsive Neural Networks with Time-Varying Delays

Neural Processing Letters - In this article, we investigate the robust passivity and stability analysis of uncertain complex-valued impulsive neural network (UCVINN) models with time-varying...     

Object - oriented databases for quality function deployment and Taguchi methods

In this paper, Object-Oriented Databases are proposed as being a new tool for Quality Function Deployment (QFD) and Taguchi Methods. Quality Function Deployment can help identify key product or process concerns with respect to customer requirements. Taguchi Methods, such as using Design of Experiments (DOE), can help identify what product or process relationships truly exist, their relative strengths and the nature of the relationship. In order to implement these two powerful quality control techniques for assuring that customer requirements are consistently met, a well organized information is required. Object-Oriented Databases are able to store, organize and manipulate both, the customer requirements and the product information for the performance of QFD and DOE.     

Influence of crystallite size on the hardness and fatigue life of steel samples coated with electrodeposited nanocrystalline Ni-W alloys

The present work deals with the effect of crystallite size on the hardness and fatigue life of steel samples coated with electrodeposited nanocrystalline Ni-W alloys. The Ni-W alloys were electrodeposited on steel samples at four different current densities (0.05, 0.10. 0.15 and 0.20 A/cm²) and at a temperature of 75 °C. The crystallite size of the deposit reduced (from 40 to 13 nm) with an increase in current density (from 0.05 to 0.20 A/cm²) due to an increase in the tungsten content (from 0.72 to 9.33 at.%). Ni-W alloy containing 9.33 at.% W and having a crystallite size of 13 nm exhibited the maximum hardness of 638 HV. The alloys, with the crystallite size in the range 40-13 nm, followed the direct Hall-Petch relation, i.e. hardness increased with a reduction in the crystallite size. The coated samples exhibited inferior fatigue lives compared to uncoated samples. This may be attributed to the presence of tensile residual stresses and inherent microcracks in the coatings. Among the specimens coated with Ni-W alloys, as the crystallite size decreased, the fatigue life of the specimen increased owing to the increase in hardness values.