Mobile Ad Hoc Networks (MANETs) have inherently dynamic topologies. Due to the distributed, multi-hop nature of these networks, random mobility of nodes not only affects the availability of radio links between particular node pairs, but also threatens the reliability of communication paths, service discovery, even quality of service of MANETs. In this paper, a novel Markov chain model is presented to predict link availability for MANETs. Based on a rough estimation of the initial distance between two nodes, the proposed approach is able to accurately estimate link availability in a random mobility environment. Furthermore, the proposed link availability estimation approach is integrated into Max-Min d-clustering heuristic. The enhanced clustering heuristic, called M4C, takes node mobility into account when it groups mobile nodes into clusters. Simulation results are given to verify the approach and the performance improvement of clustering algorithm. It also demonstrates the adaptability of M4C, and shows that M4C is able to achieve a tradeoff between the effectiveness of topology aggregation and cluster stabilities. The proposed algorithm can also be used to improve the availability and quality of services for MANETs.

With the consideration of real-time stream processing technology, it's important to develop high availability mechanism to guarantee stream-based application not interfered by faults caused by potential anomalies. In this paper, we present a novel online prediction technique for predicting some anomalies which may occur in the near future. Concretely, we first present a value prediction which combines the Hidden Markov Model and the Mixture of Expert Model to predict the values of feature metrics in the near future. Then we employ the Support Vector Machine to do anomaly identification, which is a procedure to identify the kind of anomaly that we are about to alarm. The purpose of our approach is to achieve a tradeoff between fault penalty and resource cost. The experiment results show that our approach is of high accuracy for common anomaly prediction and low runtime overhead.

The location-based routing protocol has proven to be scalable and efficient in large wireless sensor networks with mobile sinks. A great challenge in location-based routing protocols is the design of scalable distributed location service that tracks the current locations of mobile sinks. Although various location services have been proposed in the literature, hierarchical-based location services have the significant advantage of high scalability. However, most of them depend on a global hierarchy of grids. A major disadvantage of this design is that high control overhead occurs when mobile sinks cross the boundaries of the top level grids. In this paper, we introduce Hierarchical Ring Location Service (HRLS) protocol, a practical distributed location service that provides sink location information in a scalable and distributed manner. In contrast to existing hierarchical-based location services, each sink in HRLS constructs its own hierarchy of grid rings distributively. To reduce the communication overhead of location update, sinks utilize the lazy update mechanism with their indirect location. Once a sensor node detects a target, it queries the location of a sink by sending request packets in eight directions. HRLS is evaluated through mathematical analysis and simulations. Compared with a well-known hierarchical-based location service, our results show that HRLS provides a more scalable and efficient distributed location service in scenarios with various network size, sink mobility and increasing number of source nodes.

The proliferation of many-core architectures has led to the explosive development of parallel applications using programming models, such as OpenMP, TBB, and Cilk/Cilk++. With increasing number of cores, however, it becomes even harder to efficiently schedule parallel applications on these resources since current many-core runtime systems still lack effective mechanisms to support collaborative scheduling of these applications. In this paper, we study feedback-driven adaptive scheduling based on work stealing, which provides an efficient solution for concurrently executing a set of applications on many-core systems. To dynamically estimate the number of cores desired by each application, a stable feedback-driven adaptive algorithm, called SAWS, is proposed using active workers and the length of active deques, which well captures the runtime characteristics of the applications. Furthermore, a prototype system is built by extending the Cilk runtime system, and the experimental results, which are obtained on a Sun Fire server, show that SAWS has more advantages for scheduling concurrent parallel applications. Specifically, compared with existing algorithms A-Steal and WS-EQUI, SAWS improves the performances by up to 12.43% and 21.32% with respect to mean response time respectively, and 25.78% and 46.98% with respect to processor utilization, respectively.