International Journal of Cloud Computing (9 papers in press)
A Framework for Integrating Wireless Sensors and Cloud Computing
by Mohammad Jassas, Qusay Mahmoud
Abstract: Wireless sensors generate a large volume of data that require a highly scalable framework that enables storage, processing, and analysis. Cloud computing technology can provide unlimited storage in addition to a flexible processing infrastructure, allowing for the management and analysis of vast amounts of sensor data. This paper presents a framework for integrating wireless sensors and cloud computing. This framework can provide scalability and high availability for applications that use wireless sensors. Moreover, this cloud-based framework is designed to immediately make decisions based on real-time sensor and historical data, and a list of sensor and user policies are defined by the system administrator. In order to evaluate the framework performance after applying scalability and availability techniques, a load testing environment was built in the cloud to simulate a large number of virtual users. This environment was created in order to examine the quality of the services as provided by Windows Azure. The results have shown that the use of scalability techniques can significantly increase availability and performance.
Keywords: cloud computing; Wireless Sensor Networks; Raspberry Pi; Windows Azure;
Design and Evaluation of a Fuzzy Cloud Controller
by Bill Karakostas, Dimitris Kardaras
Abstract: The management of Cloud deployments is still largely the responsibility of system administrators. Introducing autonomy in Cloud management would entail, amongst other things, the ability for automated Cloud manager systems to scale up or down the number of deployed virtual machines, and/or deploy machines of different types to meet performance and other SLAs. However, the workload of Cloud deployments can exhibit high variability over short time periods. This creates the necessity of introducing autonomic behaviour in the resource management function of the Cloud deployment that makes decisions in real time so to optimise cost and/or performance. In this paper, we employ fuzzy logic to design a cloud controller whose scaling policies emulate the reasoning of human Cloud operators, and take into account the current system load and the system load rate of change. The fuzzy controller's performance is compared and evaluated against Amazons AWS auto-scaling policies. Results indicate the ability of the proposed fuzzy controller to adapt better to changing workloads by provisioning virtual machines appropriately to match the rate of system load change. Additionally, fuzzy rules are more intuitive than auto-scaling policies and thus easier to understand and modify by human users.
Keywords: autonomic Cloud controller, Cloud infrastructure management, fuzzy controller, fuzzy rule, Juju, Linux Virtual Container.
Competitive infrastructure up-gradation in federated cloud
by Subhasis Thakur
Abstract: A federated cloud is a collaboration among a set of Cloud Providers(CP) and it acts as a connected network. In a federated cloud, the virtual network requests are mapped to the resources belonging to multiple CPs. Despite being a collaboration, an individual CP gets better utility if the usage of its own part of the network increases. In this paper, we study the problem of upgrading the federated cloud.We only consider the connectivity improvement among the CPs. The usage of the part of the network owned by a CP depends on its connectivity with the rest of the network. Better the connectivity better the usage. A CP will be interested in improving the network in such a way that such improvement makes its part of the network more accessible from other clouds. At the same time, being competitive, it will avoid any improvement that facilitates better accessibility among other CPs. In this paper we study how the CPs decide on such improvement by themselves. We model this problem as a non-cooperative game. We study the existence of the pure Nash equilibrium and study the complexity to find the pure Nash equilibrium. Further, using price of anarchy, we demonstrate the effect of the selfish choices made by the CPs on the overall performance of the network.
Keywords: Federated cloud, Nash equilibrium, connectivity augmentation
Fuzzy Logic Based VM Selection Strategy for Cloud Environment
by Mohammad Alaul Haque Monil, M. Rashedur Rahman
Abstract: Server consolidation allows data centers to increase resource utilization and reduce electric power consumption by gathering multiple virtual machines into physical servers. When a server is found to be overloaded or underloaded, a list of Virtual Machines (VMs) must be selected to migrate. VM selection is an important task as the energy consumption depends on it. A number of VM selection strategies have been proposed in the literature. Each VM selection method has its distinguished characteristics. For example, one VM selection method has low energy consumption but high SLA violation and vice versa. As different VM migration approaches offer distinct advantages, fuzzy logic has been explored and applied in this research to innovate a new VM selection method which will provide a balanced and better result in all Quality of Service (QoS) measures. With this goal, we designed the Fuzzy VM selection using three different metrics. We have devised the membership function and inference rule based on the data provided by real Cloud workload data, e.g., PanetLab. We implement our novel Fuzzy VM selection method in CloudSim toolkit and compare the performance with the existing methods. Simulation results demonstrate that Fuzzy VM selection method provides a balanced performance by trading off power and other QoS parameters compared to other VM selection strategies found in literature.
Keywords: VM Placement; VM Selection; Virtual Machine Consolidation; Overload Detection; Fuzzy Logic.
Mathematical Model of Security Approaches on Cloud Computing
by Zico Mutum
Abstract: Over the years, there is a significant advance in cloud computing technology. It has grown from being a promising business concept to one of the fast growing sector of the information technology organizations. However, customers are still reluctant to deploy their business in the cloud. Security is one of the key factor which hampers the growth of cloud computing. The exponential growth of the Internet users has also lead to a significant growth of threats. Mathematical modeling plays as an important tool in analyzing and mitigating possible threats or attacks on cloud computing. This paper focuses on mathematical models on security issues arising from the usage of cloud services. We propose a model for detecting possible threats on cloud system using probability process.
Keywords: Cloud Computing; Security; Modeling; Simulation; Probability; Threats; Distribution Function; Poisson; Exponential; Infection; Transition matrix.
A Scalable Resource Provisioning Scheme for the Cloud using Peer to Peer Resource Discovery and Multi Attribute Utility Theory
by Rayhanur Rahman, Kazi Sakib
Abstract: Serving large number of users without compromising service availability and performance is key to the success of the Cloud. A fundamental challenge in building such services is incorporating scalability and fail safe techniques for discovering and provisioning of resources. As Peer to Peer (P2P) architectures are invincible to these setbacks, the work proposes a P2P based resource discovery and provisioning method for the Cloud. It first addresses the multi attribute data publishing and the range querying inability of existing Distributed Hash Table (DHT) based P2P schemes and proposes an attribute hub based discovery of provisioning information. Focused on that, a decentralized resource provisioning model is proposed using Multi Attribute Utility Theory methods. The simulation shows that the proposed approach is 44:24%and 45:81% faster than the centralized and DHT based approaches respectively in case of multidimensional range querying. It also shows the lesser number of Service Level Objective (SLO) violations and migrations which are about 24:11% and 33:43% respectively.
Keywords: Cloud; P2P; MAUT; Resource Provisioning; Resource Discovery.
Cloud Computing Resource Allocation Taxonomies
by Fabio Lopez-Pires, Benjamín Barán
Abstract: Cloud computing datacenters dynamically provide millions of virtual machines in actual cloud computing markets. Several challenging problems have to be addressed towards an efficient resource management of these cloud computing infrastructures. In the context of resource allocation, Virtual Machine Placement (VMP) is one of the most studied problems with several possible formulations and a large number of existing optimization criteria, considering solutions with high economical and ecological impact. Based on systematic reviews of the VMP literature, this work presents novel taxonomies in order to: (1) understand different possible environments where a VMP problem could be studied from both provider and broker perspectives in different deployment architectures, (2) identify existing approaches for the formulation and resolution of the VMP as an optimization problem and (3) present a detailed view of the VMP problem, identifying research opportunities to further advance in this area.
Keywords: Resource Allocation; Virtual Machine Placement; Cloud Computing;rnCloud Brokerage; Cloud Infrastructure; Optimization; Datacenters; Taxonomy.
Design and Implementation of a Framework for Provisioning Algorithms as a Service
by Abdullah Qasem, Qusay Mahmoud
Abstract: Designing, implementing and executing algorithms have become a relevant and important element in various fields. Public users and data researchers are interested in analyzing and interpreting data with shorter execution time and higher performance. Cloud computing is an environment that provides scalable and high-end virtual resources to achieve high quality services.
This paper presents the design, implementation and evaluation of a framework for provisioning algorithms as a service in the cloud. This framework introduces solutions to help clients overcome different concerns and difficulties, such as looking for an appropriate algorithm, understanding algorithm source code, installing and configuring specific libraries, and achieving high algorithmic performance. The framework provides clients the possibility to discover available algorithms and/or deploy new algorithms over multiple scalable platforms. It also allows clients to analyze data, compare results, and measure algorithms performance.
A prototype implementation of the framework has been developed to demonstrate the feasibility of the solution. Evaluating results demonstrate that providing multiple scalability models and high-end web servers will improve algorithm performance and achieve availability and reliability using the framework.
Keywords: Algorithms as a Service; Amazon Web Services; Cloud Computing; Software as a Service; Scalability Models; Sequential and Parallel Algorithms.
Task Scheduling and Virtual Resource Optimizing in Hadoop YARN-based Cloud Computing Environment
by Frederic Nzanywayingoma
Abstract: we are living in the data world where a high volume of data is changing the way things used to be in traditional IT industry. Big Data is being generated everywhere around us at all times by cameras, mobile devices, sensors, and software logs with large amount of data in units of hundreds of terabytes to petabytes. Therefore, to analyze these massive data, new skills, intensive applications and storage clusters are needed. Apache Hadoop is one of the most recently popular tools developed for big data processing. It has been deployed by many giant companies to stream large files in big datasets. The main purpose in this paper is to analyze different scheduling algorithms that can help to achieve better performance, efficiency and reliability of Hadoop YARN environment. We describe some task schedulers which consider different levels of Hadoop such as FIFO (First In First Out) scheduler, fair scheduler, delay scheduler, deadline constraint scheduler, dynamic priority scheduling, capacity scheduler, and we analyze the performance of these widely used Hadoop task schedulers based on the following elements: makespan; turnaround time; and throughput. A reliable scheduling algorithm is suggested which can work efficiently in Hadoop environments. To conclude this paper, the experimental results were given.
Keywords: Hadoop; MapReduce; Task Scheduling; YARN; HDFS; JobTracker; TaskTracker.