Summary Column‐stores perform significantly better than row‐stores on analytical workloads such as those found in data warehouses, decision support, and business intelligence applications. As mainstream data warehouses are growing into multi‐terabyte range, decision support queries should be processed in parallel to achieve adequate performance. Researchers of the column‐oriented join queries assume an unlimited reserve of main memory and focus on minimising execution time. However, some analytics require a large amount of memory to calculate intermediate results, and some interactive analytics require a fast initial response time even though queries need to process a large amount of data. Motivated by these requirements, we present a new progressive parallel star algorithm for main memory column‐stores known as “Nimble Join.” Equipped with multi‐attribute array table and a novel progressive materialisation technique, Nimble Join requires half the memory space and has two times faster initial response whilst having comparable execution time to the existing algorithm.

Summary Sensors are widely used in the field of manufacturing, railways, aerospace, cars, medicines, robotics, and many other aspects of our everyday life. There is an increasing need to capture, store, and analyse the dynamic semi‐structured data from those sensors. A similar growth of semi‐structured data in the modern web has led to the creation of NoSQL data stores for scalability, availability, and performance, whereas large‐scale data processing frameworks for parallel analysis. NoSQL data store such as MongoDB and data processing framework such as Apache Hadoop has been studied for scientific data analysis. However, there has been no study on MongoDB with Apache Spark, and there is a limited understanding of how sensor data management can benefit from these technologies, specifically for ingesting high‐velocity sensor data and parallel retrieval of high volume data. In this paper, we evaluate the performance of MongoDB sharding and no‐sharding databases with Apache Spark, to identify the right software environment for sensor data management.

Summary The efficient use of energy is essential to address concerns of cost and sustainability. Many data centers contain MapReduce clusters to process Big Data applications. A large number of machines and fault tolerance capabilities make MapReduce clusters energy inefficient. In this paper, we present a Configurator based on performance and energy models to improve the energy efficiency of MapReduce systems. Our solution is novel as it takes into account the dependence of the performance and energy consumption of a cluster on MapReduce parameters. While this dependence is known, we are the first to model it and design a Configurator to optimize these parameter settings for maximizing the energy efficiency of MapReduce systems. Our empirical evaluations show that the Configurator can result in up to 50% improvement in the energy efficiency of typical MapReduce applications in two architecturally different clusters.

Every database systems needs to employ searching algorithms to locate and retrieve data. With the proliferation of NoSQL databases, there is a need to design search algorithms that are optimised for the non-relational files and record structures. We propose several search algorithms for documentbased databases. The algorithms were designed with parallelism in mind, considering many of the NoSQL databases have very large volume of data. The algorithms were implemented and extensively tested on MongoDB and Apache Spark environment. The test results shows a promising performance of our proposed algorithms.

Energy efficiency is a major concern in today's data centers that house large scale distributed processing systems such as data parallel MapReduce clusters. Modern power aware systems utilize the dynamic voltage and frequency scaling mechanism available in processors to manage the energy consumption. In this paper, we initially characterize the energy efficiency of MapReduce jobs with respect to built-in power governors. Our analysis indicates that while a built-in power governor provides the best energy efficiency for a job that is CPU as well as IO intensive, a common CPU-frequency across the cluster provides best the energy efficiency for other types of jobs. In order to identify this optimal frequency setting, we derive energy and performance models for MapReduce jobs on a HPC cluster and validate these models experimentally on different platforms. We demonstrate how these models can be used to improve energy efficiency of the machine learning MapReduce applications running on the Yarn platform. The execution of jobs at their optimal frequencies improves the energy efficiency by average 25% over the default governor setting. In case of mixed workloads, the energy efficiency improves by up to 10% when we use an optimal CPU-frequency across the cluster.

Reverse Nearest Neighbour queries is known for the heavyweight algorithm that makes it difficult to be implemented in mobile devices due to high computations needed to verify the objects. Since the rapid development of mobile devices' hardware and also the new lightweight approach in solving the Reverse Nearest Neighbour problem, this problem can be solved in mobile devices. In this paper, we implemented the Contact Zone algorithm to create bichromatic reverse nearest neighbour region for a specific query point on mobile devices. The model is developed in a closed wireless network and various types of mobile devices with different hardware specifications are used. Our experiments show that RNN queries by region can be solved in mobile devices and different mobile CPUs do not give significant performance in processing the queries.

The disproportionate growth of the number of vehicles compared to the available and even growing road infrastructure results in severe traffic congestion in metropolitan areas, causing tremendous tangible and consequential losses in all sectors, especially in fast developing Asian economies. Bangkok is one of the cities where traffic congestion is a crucial problem. Bangkok is an old city where high density residential areas are combined with ineffective road network. The Traffic Information Systems (TISs) can play a significant role towards improving traffic congestion problems. In this paper, we present the analytical results of our quantitative research which studies various aspects related to the knowledge-based TIS. The analysis of factors that affect Bangkok's traffic as perceived by drivers, or what we call influential factors (IF), along with the impact level of each IF are reported. In addition, the potential use of social networks for TIS is also discussed. This paper also highlights the success of using the social network to reach out the massive number of people who provide feedback and thus dramatically increase the usefulness of this information for the TIS. The reported results not only strongly confirm our selection of influential context attributes in our previous study, but also confirm the feasibility of using traffic-related data from social networks in Bangkok as context attributes for our framework. The Weight Mean Score method of influential factors presented in this paper can be further enhanced and become the metric to improve our proposed framework. Our analysis and result can also guide the design of the knowledge-based Traffic Information Systems. Although the research study was done based on Bangkok data, it can also be applicable to other cities that have similar road infrastructure problems as Bangkok.

Recent advances in mobile computing coupled with the widespread availability of inexpensive mobile devices are the key motivating factors for the development of mobile health monitoring systems. However, to leverage the full potential of such systems for continuous and real time monitoring, there ar

SUMMARY Common reverse nearest neighbor queries in spatial database run in an inefficient way because they need to check a query result with almost every nearest neighbor. This wastes many time and resources, making this approach unsuitable for mobile computation. Instead of using the neighbors as candidates for the query result, a region approach can be used to answer the query. By using this approach, any objects located in the region will be considered candidate results for the query. To reduce the cost of creating the region, we introduce the concept of a contact zone, a method that can identify the right region generator points without having to process the whole points in the space, hence make reverse nearest neighbor queries by region possible to be run in mobile devices. Copyright © 2013 John Wiley & Sons, Ltd.

What if physical artifacts or devices can be aware of each others' physical presence and location, and interact with each other without user intervention or to enable innovative applications? We propose a model for devices to manage awareness of each other, extending a spatial model of interaction previously used in virtual environments. While there has been previous work on cooperative artifacts, our model is unique in introducing a quantitative technique. Moreover, our model is novel in adding to proximity-based interactions among devices the concepts of the following: 1) aura collision types based on relative locations of devices and 2) multiple (adjustable) levels of awareness and concealment measures so that each device can control how much it wants to be aware of others and how much it wants to be concealed from others. Our model is general and supports awareness of devices in (sufficiently) close physical proximity and the right aura sizes. Such devices' awareness of each other facilitates or triggers interaction, and normally precedes interaction among devices (as in human communication). Our model has numerous applications, from smart soft-toy features to proximity-triggered data exchanges.

Data quality is an important component of modern scientific discovery. Many scientific discovery processes consume data from a diverse array of resources such as streaming sensor networks, web services, and databases. The validity of a scientific computation's results is highly dependent on the quality of these input data. Scientific workflow systems are being increasingly used to automate scientific computations by facilitating experiment design, data capture, integration, processing, and analysis. These workflows may execute in grid or cloud environments, and if the data produced during workflow execution is deemed unusable or low in quality, execution should stop to prevent wasting these valuable resources. We propose an approach in the Kepler scientific workflow system for monitoring data quality and demonstrate its use for oceanography and bioinformatics domains.

Grid computing provides opportunities for scientists to make breakthrough discoveries through new ways to conduct experiments that were not possible in the past. The growing availability of grid infrastructure to the wider scientific community has increased the amount of data produced by these experiments. Yet, without proper management for storing, retrieving and disseminating the results, potential benefits of grid computing may not be fully realized.

In the data mining field, association rules are discovered having domain knowledge specified as a minimum support threshold. The accuracy in setting up this threshold directly influences the number and the quality of association rules discovered. Often, the number of association rules, even though large in number, misses some interesting rules and the rules' quality necessitates further analysis. As a result, decision making using these rules could lead to risky actions. We propose a framework to discover domain knowledge report as coherent rules. Coherent rules are discovered based on the properties of propositional logic, and therefore, requires no background knowledge to generate them. From the coherent rules discovered, association rules can be derived objectively and directly without knowing the level of minimum support threshold required. We provide analysis of the rules compare to those discovered via the a priori.

Due to its ability to provide high-performance computing environment, the grid has become an important infrastructure to support eScience. To utilise the grid for parameter sweep experiments, workflow technology combined with tools such as Nimrod/K are used to orchestrate and automate scientific services provided on the grid. As parameter sweeping over a workflow needs to be executed numerous times, it is more efficient to execute multiple instances of the workflow in parallel. However, this parallel execution can be delayed as every workflow instance requires the same set of resources leading to resource competition problem. Although many algorithms exist for scheduling grid workflows, there is little effort in considering multiple workflow instances and resource competition in the scheduling process. In this paper, we proposed a scheduling algorithm for parameter sweep workflow based on resource competition. The proposed algorithm aims to support multiple workflow instances and avoid allocating resources with high resource competition to minimise delay due to the blocking of tasks. The result is evaluated using simulation to compare with an existing scheduling algorithm.

We propose a conceptual framework to describe and understand mobile information environments. In our proposal, such an environment can be categorised into different abstraction levels or communities: user level, workflow level, device level and architecture level. A hospital-based example is then used as an illustration for the proposed framework.

Recently, workflow has been employed in pervasive computing systems to orchestrate and provide services to the users. Since pervasive computing systems emphasise context-awareness and adaptability, these two features must be included in the workflow mechanism in order to utilise workflow in pervasive environment. In this paper we present various existing approaches to adaptation in context-aware workflow. These approaches are compared based on the common adaptation characteristics identified from our study. Lastly, we discuss and point out the future challenges of predictive adaptation for context-aware workflows.

Embedded with sensors and appropriate computational entities, a road intersection can be viewed as a pervasive computing environment. The crash rate in road intersections demonstrates the need for a fast and accurate collision detection system. We suggest that an intersection collision detection system should be able to adapt to different types of intersections for faster collision detection. Moreover, a real-time application-level communication protocol to warn affected drivers is required. An intersection agent that takes vehicular status information from vehicle agents and learns, detects and warns collisions at a road intersection is proposed. The issues, challenges, and cost of a multiagent collision avoidance system are discussed. A communication protocol that is designed specifically with intersection safety in mind is presented here.

We present AgentUDM a mobile agent based support infrastructure for ubiquitous data mining (UDM). The focus of AgentUDM is to address the issues of resource-constraints and disconnections with respect to UDM.

One major drawback in using a client server system in the mobile environment is the redundancy of data transmission in the event of connection failure. In this paper, we present a novel mobile agent architecture which allows minimum redundancy of data transmission in a client server database connection in a frequent disconnections environment. The system supports disconnection events caused by network failure or client power failure. Unlike current approaches, our model uses agent messaging to transfer the data. The messaging reduces the amount of migrations that the mobile agents performed. The architecture was implemented and tested in the Aglets platform.

Bayesian belief networks have been used widely to solve many decision problem that involve uncertainty. One major advantage of this approach compared with other reasoning tools is its semantic richness in describing the decision process. Some inference algorithms for carrying out the reasoning process exist, but they are known to be computationally expensive. Hence, they require optimisation to make them practical. This paper proposes two optimisation techniques for Bayesian belief networks. These optimisation techniques were investigated for information retrieval applications, but can also be applied to different applications outside the information retrieval area.