Papers by Author: Bin Zhang

Abstract: Software aging threatens to the reliability of the software and computer system seriously, and has become the main bottleneck restricting the development of software reliability. As most of user behavior has certain regularity, so it can be forecast. This paper proposed a prediction method based on the Frequent-User-Access-Patterns-Tree-With-Time-And-Aging (FUAP-Tree). And this method compressed the user behavior into a FUAP-Tree to store the Web log, and then predicted the user behavior using pattern matching by traversing the sub-trees of the FUAP-Tree. The load of the Web application software could be obtained when user behavior is known. So we could predict the Web application software aging. The conclusion we got by the experiment of comparing with another method is that this method got a more effective and accurate result.

Abstract: Although SaaS is a new application model, the phenomenon of software aging still exists, which will cause service performance decline ceaselessly, even worse, service failure. In order to ensure SaaS performance, we need to estimate the degree of software aging accurately, so as to provide a basis to the resource allocation strategy. In this paper we describe the process of calculating software aging degree, and build a performance metrics system according to gray correlation degree analysis, then calculate SaaS software aging degree based on fuzzy evaluation. Finally, we verify the effectiveness of our measurement method of software aging in cloud by experiments.

Abstract: Two different laminated composites with submicron-scale grain size and strong interface bonding toughness, Cu/Al and Cu/Cu, were fabricated by cold-roll bonding at ambient temperature, and then annealing of the laminated composites was conducted to get different interface bonding toughness. It was found that a better strength-plasticity combination for the laminated composites could be obtained through stronger interface bonding toughness, which effectively delayed the onset of plastic instability and premature local necking of the material. Uniform elongation of both Cu/Al and Cu/Cu laminated composites was enhanced compared with that of the cold-rolled Cu. At the same strength level, plasticity of the Cu/Cu laminated composite is better than that of the Cu/Al one and that of the cold-rolled Cu. Mechanisms of plasticity instability and fracture of the laminated composites were evaluated.

Abstract: Traditional load balancing strategy is based on the assumption that each node is valid ,in practice, with the growth of usage time, the failure rate of nodes will gradually increase, so we introduce software aging rejuvenation strategy into load balancing. We propose an algorithm to solve this problem. It is different from other algorithm that the function of collecting load information from real servers is moved to back-end in this algorithm, load balancer assigns tasks according to the calculation results, it could reduce the load of load balancer own. At the same time, in order to prevent unexpected event due to computer itself (e.g. cpu utilization rate sudden change), we also take into account decay factor. From the experiment result we can see that compared with non software aging rejuvenation strategy in load balancing, this strategy could balance the load of all the nodes more effectively.

Abstract: An in-situ testing system for thermal-mechanical fatigue of thin metal lines was setup inside a dual-beam focused ion beam (FIB)/scanning electron microscope (SEM) system. Alternating currents (AC) were applied to narrow Au lines 200-nm-thick through nanomanipulator needles. Preliminary results show that severe thermal-mechanical fatigue damage can be generated by the action of the applied AC. The in-situ recording of the evolution of the damage has been carried out and the possible mechanism of the thermal-mechanical fatigue damage in the Au lines resulted from the joule heating was discussed.

Abstract: Fatigue tests of nanometer-thick Cu films as deposited and annealed in vacuum were conducted under constant load ranges at room temperature. Fatigue strengths of the Cu films, which is defined as the critical load range being able to cause crack initiation within 106 cycles, are determined. The experimental results show that fatigue strength increases with decreasing film thickness. Fatigue cracking behaviors were characterized by electron microscope. It is also found that fatigue cracking resistance is dependent on film thickness and increases with decreasing film thickness. Size effects on fatigue properties of the nanometer-thick Cu films are discussed.