Earticle

This paper presents multidimensional aggregation query processing algorithm in cloud computing system. The existing cloud computing research work in the MapReduce calculation framework lacks effective support to the aggregation of multi-dimensional data. On the other hand, the use of MapReduce computing framework needs to start large computing nodes, and costs huge amounts of energy. For the above problems, this paper presents multidimensional aggregation operation scheme in the cloud computing system, through two layers of index structure to reduce the query computing nodes and Calculation of aggregation operation in single computing nodes. This paper gives an algorithm framework using two layers of index structure to process multidimensional aggregation query, and proposed priority mode of performance and multi-dimensional aggregation algorithms under low-power mode in this framework. In two modes of multidimensional aggregation algorithm proposed query allocation problem, and proves that the two modes query allocation problems are NP-completing problem. This paper presents approximation algorithm of two NP-completing problem and proves the approximate ratio of two approximation algorithms. Theoretical analysis and simulation results prove the validity of the multidimensional aggregate query plan.

Wireless Sensor Network is an emerging new technology extensively used for large-scale real-time data processing due to its characteristics and is used in application areas such as in civilian and military domains. These networks are vulnerable to several security attacks. In this paper, we emphasize on Sybil attack, which is a harmful threat to sensor networks and proposed an algorithm for Sybil attack detection based on Time difference of Arrival (TDOA) localization method, which detects the malicious behavior of the head node and member nodes in a cluster based network.

In large-scale grid platforms, providing fault-tolerance for users is always a challenging task because of the uncertainty of network resources. In this paper, we present an intelligent agent based meta-scheduler, which is aiming at improving the fault-tolerance of grid systems when running user’s application. The proposed meta-scheduler is designed as an extendable framework, which allows plugging in multiple scheduling policies for deal with different scenarios. The agent-based scheduling framework enables grid systems to deploy their local schedulers in a flexible manner. Extensive experiments are conducted to investigate the performance of the proposed meta-scheduler, and the results show that it is effective to provide enhanced dependability for grid users, especially when the system is across multi-organization.

Wireless sensor network (WSN) is a collection of densely deployed sensor nodes. These nodes are prone to attack and also resource constrained. Resource accountability is also required for Security implementation in WSN. Some researchers proposed different methods, algorithms and frameworks for WSN security implementation. SPINS is a framework for implementing overall security in WSN using SNEP (Sensor Network Encryption Protocol) and μTESLA (the “micro” version of the Timed, Efficient, Streaming, Loss-tolerant Authentication Protocol) protocol. Existing research works reveal that SNEP employed RC5 encryption algorithm for WSN. This paper analyzed efficient encryption algorithm XTEA for SNEP & evaluate that it is better as compare to RC5 in terms of energy, storage and time.

Financial time series prediction is regarded as one of the most challengingtasks due to the inherent noise and non-stationality of the data. This paperproposed a two-stage financial time series prediction approach hybridizingsupport vector regression (SVR) with hierarchical clustering (HC). By averaging the variables within the clusters obtained from hierarchical clustering, we define super predictors and use them as the input variables of the SVRforecasting model. Although averaging is a simple technique, it plays animportant role in reducing variance. To evaluate the performance of theproposed approach, the Shanghai-Shenzhen 300 index is used as illustrativeexample. The experimental results show that the proposed approach outperforms both the SVR model with principal component analysis and the SVRmodel with genetic algorithms in average prediction error and predictionaccuracy.

With the increasing popularity of cloud computing, more and more sensitive or private information has been outsourced onto the cloud server. For protecting data privacy, sensitive data usually has to be encrypted before outsourcing, which makes traditional search techniques based on plaintext useless. In response to the search over encrypted data, searchable encryption is a good solution in Information Security. However, most of existing searchable encryption schemes only support exact keyword search. That means they don’t support searching for different variants of the query word, which is a significant drawback and greatly affects data usability and user experience. Recently, a fuzzy keyword search scheme proposed by some researchers aims at addressing the problems of minor typos and format inconsistence but couldn’t solve the problem above. In this paper, we formalize the problem of semantic keyword-based search over encrypted cloud data while preserving privacy. Semantic keyword-based search will greatly improves the user experience by returning all the documents containing semantically close keywords related to the query word. In our solution, we use the stemming algorithm to construct stem set, which reduces the dimension of index. And the symbol-based trie is also adopted in index construction to improve the search efficiency. Through rigorous privacy analysis and experiment on real dataset, our scheme is secure and efficient.

The Congestion in wireless sensor networks occurs when packets arrive at a fully loaded buffer of a sensor node. Due to lack of space in the buffer, packets are dropped. It also leads to packet delay, retransmission of packets, reduced QoS and throughput. A hybrid of rate control and resource control algorithms can be used to control congestion considering the priority of the packets. Congestion is detected when the service time is greater than the inter-arrival time of packets. The inter-arrival time can be increased by a certain factor computed as congestion degree, thus reducing the rate of sending packets. The congestion degree which is the ratio of service time to inter-arrival time can be sent to the sources. The sources can then adjust the rate of sending packets accordingly to control congestion in the upstream nodes. At the time of congestion, high priority packets are sent using multiple paths after increasing the time between generations of packets thus decreasing the rate of sending packets. Low priority packets are sent using a single path after decreasing the packet sending rate thus increasing the time between sending consecutive packets. The method also serves as a hybrid of congestion control and congestion avoidance.

In cloud computing environment, there are a large number of users which lead to huge amount of tasks to be processed by system. In order to make the system complete the service requests efficiently, how to schedule the tasks becomes the focus of cloud computing Research. A task scheduling algorithm based on PSO and ACO for cloud computing is presented in this paper. First, the algorithm uses particle swarm optimization algorithm to get the initial solution quickly, and then according to this scheduling result the initial pheromone distribution of ant colony algorithm is generated. Finally, the ant colony algorithm is used to get the optimal solution of task scheduling. The experiment simulated on CloudSim platform shows that the algorithm has good effect in real-time performance and optimization capability. It is an effective task scheduling algorithm.

The recent advances in cloud computing set up well established research in Distributed computing, virtualization, web services, utility computing, have offered many benefits in scalability, cost and efficiency for cloud service users. These advantages are further expected to fulfill the demands of cloud users for cloud services efficiently. This brings the problem of fault tolerance, scalability, efficiency, high availability. Central to these issues require the dynamic and efficient load balancing techniques. In this paper, an effective load balance model is presented for public cloud in which cloud is partitioned with a switch mechanism to choose different load balancing strategy for different load. Fuzzy Logic have been extensively used in various applications such as image processing, data mining, networking, etc. due to its efficient internal architecture and its compatibility to solve various optimization problems. Fuzzy base networks have been observed to produce optimal results in various combinatorial optimization problems. Another important area which provides significant results in solving optimization is the swarm intelligence approach. GSO is observed to have provided significant optimal solution in lesser iterations. In this paper The Fuzzy logic and GSO based load balancing algorithm applied to the load balancing strategy to enhance the utilization and efficiency in the public cloud environment.

Vehicular Ad-hoc Networks (VANETs) are expected to provide assistance to various applications, such as accident broadcast or emergency notification. Some kinds of applications need data to be delivered in a certain region by geocast, which means disseminating information in a specific area, instead of using a certain IP address. In this paper, we propose a data dissemination protocol for VANETs, which delivers data to all the vehicles in an interested region. We assume each vehicle is equipped with GPS and DSRC devices. Our simulation results show that this data dissemination protocol can contribute to high delivery ratio. Also, it performs well in adaptability during heavy traffic.

Due to recent advancement in wireless sensor technology and extensive requirement of scalable applications, researchers are designing models of sensor network that can support wide multiplicity of services in a single heterogeneous sensor network infrastructure. In this work, we survey the virtual sensor networks (VSNs) design architecture, principles, service provisioning, sensor cloud architecture, and various applications that are widely used in different areas of VSNs. We propose a new framework for VSNs that can support multiple services and can give a new direction to Wireless Sensor Network (WSN) community.

This paper presents a fuzzy normal estimate for mass point clouds of irregular models in reconstruction. The irregular model is complex object that some part is smooth and some parts are irregular including sharp features. Therefore, we put kNN and curvature of mass point clouds to fuzzy inference system to divide the kind of point clouds and the output of FIS can determine which part of tooth point clouds belong to. For different kinds point clouds, corresponding algorithm is given. Point clouds in the smooth area are estimated normal by PCA directly and ones in other regions of thin or sharp area are estimated by checker and attach points. This method is simpler than those complex methods used on the whole point clouds directly. The experiment results show that much time is saved and surface reconstruction is very fine than PCA and WLOP.

In this paper, a BCC-DPSO scheduling algorithm is proposed to solve multi-objective optimization problem for task scheduling on Network-on-Chip (NoC). In our proposal, the relative advantage of the solution is evaluated by calculating its efficiency using BCC model in Data Envelopment Analysis (DEA), and the referred-time method is introduced to rank the BCC-efficient solution. Moreover, a sub-swarm strategy is adopted to reduce the high computational requirement introduced by the DEA. There are four sub-swarms, each of which optimizes one of four observed metrics, namely makespan, energy, link load and workload balance. Meanwhile, the speed vector updating formulation is modified to comply with the sub-swarm strategy. By conducting comparative simulations, the results show that our proposal produces more efficient schedule solution than other multi-objective Particle Swarm Optimization (PSO).

Using mathematical method to model Web workload locality characteristics is established based on the study of entropy and coefficient of variation. Experiments show that these metrics can depict locality characteristics more properly and are much easier to use.

The purpose of this paper is concerned with a class of multi objective programming problems in which the objective functions and constrained functions are twice differentiable and containing the support functions of a compact convex set. For such mathematic programming, to deal with its dual problem, first, the concept of second order B - (p,r) - invexity is introduced. Further, the Mangasarian type second order dual model associated with the multi objective problem are formulated. Several weak, strong and strict converse dual theorems are established and proved by utilizing the new generalized convexity. The results extend and improve the corresponding results in the literature.

The demand for computing resources is growing, but the input costs of the enterprise information system construction of the traditional way is too high, and has not the high utilization of the resources. In order to solve this contradiction, the enterprise private cloud platform architecture has been put forward based on open source system. By analyzing the mainstream open source cloud projects, the basis to choose OpenStack has been given. From the perspective of classification of the enterprise personnel, combining with the components of OpenStack, this paper expounds manufacture of the virtual image and distribution, monitor, and use of the virtual machine. At last, the user interface design is also given.

Cloud computing, with its powerful capability of computing and storage and with its large-scale data resource, provides unprecedented opportunities for management and application of human resource information. How to support complex and even unpredictable human resource services as well as manipulate large scale data at the same time to provide good quality of human resource information, is one of the crucial and fundamental problems for cloud-based human resource information system. This paper proposes establishing a general framework for human resource application and services based on cloud. As a type of compact, structured, and knowledgeable human resource information representation, and publish to the user by Internet. Moreover, the case is evidenced by an application example.