Kaigui Wu

Multilevel redundancy allocation using two dimensional arrays encoding and hybrid genetic algorithm

With the popularity of multilevel design in large scale systems, reliability redundancy allocation on multilevel systems is becoming attractive to researchers. Multilevel redundancy allocation problem (MLRAP) is not only NP-hard, but also qualifies as hierarchy optimization problem. Exact method could not tackle MLRAP very well, so heuristic and meta-heuristic methods are often used to solve it. To improve the effectiveness of current algorithms on MLRAP, this paper proposes a hybrid genetic algorithm (HGA) based on the two dimensional redundancy encoding mechanism. Instead of hierarchical genotype representation, a two dimensional array is used to represent the solutions to MLRAP. Each row of the array contains the redundancy information of a certain unit in the system and each element in one row stands for the redundancy value of one element of that unit. The number of rows of this array is fixed and equals to the number of distinct units in the system. Each row of the array is an unfixed-length vector whose length depends on the redundancy of all elements of its parent unit. On top of this two dimensional arrays, a local search operator employing simulated annealing strategy is used to generate new population for the next generation instead of the traditional genetic operators. Experimental results have shown that our two dimensional arrays based HGA outperforms the state-of-the-art approaches using two kinds of multilevel system structure.

A hash-based secure interface on plain connection

This paper proposes a hash-based secure interface between two nodes on the Internet, especial between two interfaces or two web pages. Digital signatures and public-private keys are traditionally used to provide integrity and authentication. This paper proposes an alternative method which uses a shared private key and a public hash function for a message that is sent over a plain connection without losing integrity and authentication. An additional private algorithm is needed when the message has been hashed based on the message and the salt (the shared private key), and the hashed value will be re-computed with the private algorithm to produce a string named as checksum. At the other end, when a message is received with a checksum the same process is followed to produce a new checksum. If the produced checksum is equal to the received checksum, the message is legitimate. For efficiency and reliability, a timestamp and validity period is introduced to the scheme. The salt gets more salty with time included and so does the secure interface.

Monitoring Resources Allocation for Service Composition Under Different Monitoring Mechanisms

As availability of web service has become a great concern in SOA, monitoring mechanism is often deployed to detect and recover failures for service composition. While monitoring mechanism could improve the availability to an extent, it may cost more resources and increase the response time perceived by end users. To decrease the overall usage of monitoring resources, this paper proposes to select some services bringing the highest availability improvement to the composition and allocate monitors on them while leaving others unmonitored. This paper first researched two common monitoring mechanisms in service composition and analyzed their different impact on the service composition QoS values. Then continuous-time Markov chain and discrete-time Markov chain were employed to build the availability model related to the monitoring rate or service pool size according to different kind of monitoring mechanisms. Based on these models, two algorithms were proposed, for two monitoring mechanisms respectively, to allocate monitors in the composition aiming at minimizing the overall number of monitors while making sure the composition availability could meet certain requirements. The monitor allocation algorithm could be used to get the overall number of monitors and those services to monitor in different scenarios. Empirical studies results showed that it was feasible to monitor only some services in the composition to meet certain availability requirement. Monitors allocation decreased the overall number of monitors in the service composition and also decreased the mean response time comparing with the scenario that all services were monitored.

Dependable Dynamic Service Integration on Service-Oriented Peer-to-Peer Networks

For provision of dependable search and rescue capability in dynamic and unpredictable disaster areas, the networked nodes should have the ability to autonomously support and co-operate with each other in a peer-to-peer (P2P) manner to quickly configure any services available on the disaster area to deliver a real-time capability. In this paper, we present an innovative architectural approach which is able to proactively self-diagnose and self-adapt evolution occurring in provision of search and rescue capabilities in service-oriented P2P (ServP2P) networks. This architecture provides flexibility and agility in handling dynamic changes and evolution encountered in the delivery of rescue capability. The experimental results indicate that the proposed approach provides a high-level of reliability and sustainability to achieve dependable service integration for capability provision.

State-Based Search Strategy in Unstructured P2P

Efficient resource search in large-scale unstructured peer-to-peer (P2P) systems remains a fundamental challenge. In order to improve search performance, interest-based search is a good way to tackle the challenge. However, the existing interest-based search algorithms pay attention to user interest model and search history, but ignore some influence factors of improving performance. In this paper, through analysis of node’s multi-dimensional quality-of-service (QoS) for search, it is found that node state’s information is essential for improving search performance. So a State-Based Search (SBS) strategy is presented by improving Sripanidkulchai’s interest-based search model, where Node’s state evaluation is based on node’s QoS fuzzification. The SBS enhances the methods of shortcuts adding, ranking and selecting, and state’s life-cycle is predicted by grey theory. The experimental results show that the SBS can improve performance by reducing search response time and achieving load balance.

An Improved Back Propagation Neural Network Model and Its Application

Stroke is one of the most serious disease, and the incidence rate of stroke is confirmed to be related to environmental factors including temperature, pressure and humidity .In order to obtain the relationship between the incidence rate and environmental factors , we research on local daily meteorological data and stroke disease cases from January 2008 to December 2012, which is provided by the administrative department of public health and medical institutions statistics in China, then build the improved BPNN(Back propagation neural network) model to carry out data analysis and processing, obtain the weight matrix between them. It can be seen that the relationship between incidence rate and pressure is the highest degree from the value of weight matrix, and pressure is positive correlation with the incidence rate. The relationship between the temperature and incidence rate is second, and they are negative correlation. The incidence between average relative humidity and correlation is quite small. The results show that the model can be used to predict the future stroke incidence rate under various meteorological conditions, and it can play a certain role in making disease knowledge popular and providing a reference to potential patients.

Multi-Objective Memetic Algorithm for Monitoring Resources Allocation in Service Composition

While monitoring mechanism is often used to improve reliability of service composition, it also increases system cost. To optimize overall usage of resources, this paper proposed to use memetic algorithm to optimally allocate monitoring resources in service composition under certain reliability constraints. It first analyzed the reliability and cost model of service composition under monitoring mechanism and then formulated the problem as multi-objective optimization problems. After that, a multi-objective memetic algorithm (MOMA) was presented to solve this problem. This algorithm employed nondominated sorting strategy as the global search method and random walk with direction exploitation method as the local search operator. Experimental studies results showed that MOMA searched more effectively than the sensitivity-based method and other multi-objective evolutionary algorithms including NSGA II and HaD-MOEA.

Near-optimal configuration of service pool size in service composition

To improve the reliability of service composition, service pools with redundant services are often registered in advance. Increasing the number of redundant services in a pool could improve the reliability while decreasing the performance or increasing unnecessary cost. Aiming at minimizing the overall cost or response time under certain reliability constraint, this paper proposed a service pool size choosing method from the point of view of service composition. It started with the reliability and performance evaluation of single service pool with multiple services and then employed architecture-based model to obtain the overall reliability/performance model for service composition with multiple service pools. These models were used as the utility function and constraint function to optimize service pool size combination. Two sensitivity-based algorithms were proposed to solve the optimization problem, which assigned more redundant services to those service pools that have higher impact on reliability change and lower impact on performance change. Empirical studies results showed that these algorithms could find the near-optimal size combination of service pools using much less time than the exhaustive iteration method.

A dynamic reliability maintenance method for online software system

Software reliability is one of important characteristics for most systems. Due to dynamic properties of the system, the probability of number of visits to each component might change after a certain amount of period. To maintain the software system reliability, this paper proposed a dynamic reliability maintenance mechanism based on resource monitoring and redundant resources allocation. An open-source monitoring software, namely Glassbox, was employed to monitor the change of number of visits to a component. Then the path-based method was used to analyze the reliability of software system at the component level in run time. Sensitivity of each component was calculated using the path-based model. A greedy algorithm for re-allocating resources dynamically was proposed according to component sensitivity. Finally, some simulation experiments were conducted to evaluate the proposed mechanism. The results have shown that the proposed method is effective.

QoS-aware service composition in multi-network environment based on genetic algorithm

QoS-aware service composition is one of the most important technologies in SOA. The aim is to compose a new service with the best QoS under the user constraints by a set of atomic services. There are a lot of researches on this topic in recent years. But most of them did not take the effect of the multi-network environment into account and the algorithms took much time when the number of the service candidates was large. In this paper, we present a service composition model in multi-network environment and solve the problem by employing genetic algorithm (GA) within acceptable time. Our approach provides a fast way to find the best solution for QoS-aware service composition in multi-network environment. The experimental results show that our method achieves better performance than other approaches.