8th International Conference on Cloud Computing: Services and Architecture (CLOUD 2019)

July 13~14, 2019, Toronto, Canada

Accepted Papers


    Service Level Driven Job Scheduling in Multi-Tier Cloud Computing: A Biologically Inspired Approach
    Husam Suleiman and Otman Basir, University of Waterloo, Canada
    ABSTRACT
    Cloud computing environments often have to deal with random-arrival computational workloads that varyin resource requirements and demand high Quality of Service (QoS) obligations. It is typical that a Service-Level-Agreement (SLA) is employed to govern the QoS obligations of the cloud computing service providerto the client. A typical challenge service-providers face every day is maintaining a balance between thelimited resources available for computing and the high QoS requirements of varying random demands.Any imbalance in managing these conflicting objectives may result in either dissatisfied clients and po-tentially significant commercial penalties, or an over-resourced cloud computing environment that can besignificantly costly to acquire and operate. Thus, scheduling the clients’ workloads as they arrive at theenvironment to ensure their timely execution has been a central issue in cloud computing. Various ap-proaches have been reported in the literature to address this problem: Shortest-Queue, Join-Idle-Queue,Round Robin, MinMin, MaxMin, and Least Connection, to name a few. However, optimization strategiesof such approaches fail to capture QoS obligations and their associated commercial penalties. This paperpresents an approach for service-level driven load scheduling and balancing in multi-tier environments.Joint scheduling and balancing operations are employed to distribute and schedule jobs among the re-sources, such that the total waiting time of client jobs is minimized, and thus the potential of a penalty tobe incurred by the service provider is mitigated. A penalty model is used to quantify the penalty the serviceprovider incurs as a function of the jobs’ total waiting time. A Virtual-Queue abstraction is proposed tofacilitate optimal job scheduling at the tier level. This problem is NP-complete, thus, a genetic algorithm isproposed as a tool for computing job schedules that minimize potential QoS penalties and hence minimizethe likelihood of dissatisfied clients.
    KEYWORDS

    Heuristic Optimization, Job Scheduling, Job Allocation, Load Balancing, Multi-Tier Cloud Computing


    Evaluating Service Trustworthiness for Service Selection in Cloud Environment
    Maryam Amiri and Leyli Mohammad-Khanli, University of Tabriz, Iran
    ABSTRACT
    Cloud computing is becoming increasingly popular and more business applications are moving to cloud. In this regard, services that provide similar functional properties are increasing. So, the ability to select a service with the best non-functional properties, corresponding to the user preference, is necessary for the user. This paper presents an Evaluation Framework of Service Trustworthiness (EFST) that evaluates the trustworthiness of equivalent services without need to additional invocations of them. EFST extracts user preference automatically. Then, it assesses trustworthiness of services in two dimensions of qualitative and quantitative metrics based on the experiences of past usage of services. Finally, EFST determines the overall trustworthiness of services using Fuzzy Inference System (FIS). The results of experiments and simulations show that EFST is able to predict the missing values of Quality of Service (QoS) better than other competing approaches. Also, it propels users to select the most appropriate services.
    KEYWORDS

    User preference, cloud service, trustworthiness, QoS metrics, prediction


    Public Cloud Data storage and retrieval Security Model
    Sonia amamou1, Zied trifa2 and Maher khmakhem3, 1MIRACL Research Center, Sfax, Tunisia, 2ISAE Gafsa, University of Gafsa, Tunisia and 3University of King Abdulaziz, Tunisia
    ABSTRACT
    Data protection is considered to be a main facet in order to have a favorable development of the infrastructure of public cloud. This is due to the fact that it mainly indicates the well working of infrastructure especially for the cloud users and also for the cloud providers. Unfortunately, data protection has not yet been provided with the required attention. Nowadays, in order to achieve data protection, cryptographic approaches are recommended as solutions because they can grant some degree of control on data users. Nevertheless, they can reduce the efficiency and also set up an important overhead of any cloud system . In this work, we studied the storage and retrieval in the cloud, which are protection methods. Then, we evaluated their strengths and also their limitations in order to propose a new approach.
    KEYWORDS

    Public Cloud , Data Protection , storage , Cloud Security , Security Model , Data storage , Data retrieval , retrieval


    Threat Modelling for the Virtual Machine Image in Cloud Computing
    Raid Khalid Hussein, University of Southampton, United Kingdom
    ABSTRACT
    Cloud computing is one of the most attractive technology in the era of computing as its ability to reduce the cost of data processing while increasing flexibility and scalability for computer processes. Security is one of the main concerns related to the cloud computing as it hampers the organizations to adopt this technology. Infrastructure as a service (IaaS) is one of the main services of cloud computing which uses virtualization to supply virtualized computing resources to its users through the internet. Virtual Machine Image is the main component in the cloud as it is used to run an instance. There are a number of security issues related to the virtual machine image that need to be analyzed to draw future directions to secure virtual machine as being an essential component related to the cloud computing. Therefore, this paper provides a threat modelling approach to identify the threats that affect the virtual machine image. In addition, threat classification is carried out to each individual threat to find out their negative effects on the cloud computing. Potential attack was drawn to show how an adversary might exploit the weakness in the system to attack the Virtual Machine Image.
    KEYWORDS

    Cloud Security, Virtualization, Virtual Machine Image, Security Threats, Threat modeling First Section


    Secure and privacy preserving sharing of patient information in cloud-based healthcare systems
    Sanjay Sareen and Meenu Sareen, Guru Nanak Dev University, india
    ABSTRACT
    The growing need of sharing health information of patients between healthcare agencies, doctors, and other authorized persons in order to provide better care of patients has resulted in the cloud-based healthcare system. However, the sharing of health information on the cloud raises major security and privacy issues that need to address. This article introduces a framework that protects health information from unauthorized access and lets the patient as data owner decide who the authorized persons are, i.e., who the patient discloses her health information to. In this article, a framework is proposed that presents a new methodology using a combination of different techniques such as information granulation, information pseudonymization, and secret sharing scheme, allowing privacy-preserving primary and secondary use of the health records. In this model, the privacy of patient’s identity is protected so that users and providers of healthcare services do not need to trust the cloud service provider with privacy related issues. We designed and implemented different privacy preserving algorithms on the Amazon EC2 to evaluate the security and performance analysis of our proposed system. The security analysis showed that the framework is secure and protected against common intruder scenarios.
    KEYWORDS

    Pseudonymization; secret sharing scheme; cloud computing; secure data sharing; healthcare systems


    Security Considerations For Edge computing
    John M. Acken1 and Naresh K. Sehgal2, 1Portland State University, Portland, OR and 2Intel Corp, Santa Clara, CA
    ABSTRACT
    Present state of edge computing is an environment of different computing capabilities connecting via awide variety of communication paths. This situation creates both great operational capabilityopportunities and unimaginable security problems. This paper emphasizes that the traditionalapproaches to security of identifying a security threat and developing the technology and policies todefend against that threat are no longer adequate. The wide variety of security levels, computationalcapabilities, and communication channels requires a learning, responsive, varied, and individualizedapproach to information security. We describe a classification of the nature of transactions with respectto security based upon relationships, history, trust status, requested actions and resulting responsechoices. We propose that each element in the edge computing world utilizes a localized ability toestablish an adaptive learning trust model with each entity that communicates with the element
    KEYWORDS

    Edge Computing, Security, Adaptive learning, Trust model, Threats, Cloud Computing, InformationSecurity


    A New Prediction Framework For Improving the Energy Efficiency in Cloud
    Maryam Amiri and Leyli Mohammad-Khanli, University of Tabriz, Iran
    ABSTRACT
    Cloud computing relies on sharing pool of computing resources. It delivers services with availability and scalability obligations to users. On the other hand, green computing and energy efficiency are two main challenges of cloud. Therefore, resource should be allocated in a way that energy consumption is reduced and Quality of Service (QoS) dropping is avoided. Due to the point that storage is one of the biggest energy consumers, this paper focuses on the storage system in cloud. For improving energy efficiency of storage, the next access number of data blocks of disks is predicted. Based on predicted results, data blocks can be transmitted to the most appropriate disks and energy consumption is reduced. The goal of this paper is to represent a new method to predict the next access number of data blocks of disks. The proposed framework is composed of four components: fuzzy classifier, Markov chain, base prediction methods and neural network. Indeed, the main predictor component is neural network. After prediction of next state of data block, data can be distributed on the disks in a way that the number of disks is minimized due to energy consumption reduction and avoidance of Service Level Agreements (SLA) violation. The results of experiments show the high accuracy of the proposed method in comparison with the similar methods.
    KEYWORDS

    Cloud computing, Prediction framework, Energy efficiency, Neural Network, Fuzzy Classifier, Access number


    Trust Modelling for Security of IoT Devices
    Naresh K. Sehgal1, Shiv Shankar2 and John M. Acken3, 1Data Centre Group, Intel Corp, Santa Clara, CA, 2Chief Data Scientist, Maphalli, Bangalore, India and 3Portland State University, Portland, OR
    ABSTRACT
    IoT (Internet of Things), represents many kinds of devices in the field, connected to data-centers via various networks, submitting data, and allow themselves to be controlled. Connected cameras, TV, media players, access control systems, and wireless sensors are becoming pervasive. Their applications include Retail Solutions, Home, Transportation and Automotive, Industrial and Energy etc. This growth also represents security threat, as several hacker attacks been launched using these devices as agents. We explore the current environment and propose a quantitative and qualitative trust model, using a multi-dimensional exploration space, based on the hardware and software stack. This can be extended to any combination of IoT devices, and dynamically updated as the type of applications, deployment environment or any ingredients change.
    KEYWORDS

    Edge Computing, Security, Adaptive learning, Trust model, Threats, Cloud Computing, Information Security


    QOS-Driven Job Scheduling: Multi-Tier Dependency Considerations
    Husam Suleiman and Otman Basir, Department of Electrical and Computer Engineering, University of Waterloo, Canada
    ABSTRACT
    For a cloud service provider, delivering optimal system performance while fulfilling Quality of Service (QoS) obligations is critical for maintaining a viably profitable business. This goal is often hard to attain given the irregular nature of cloud computing jobs. These jobs expect high QoS on an on-demand fashion, that is on random arrival. To optimize the response to such client demands, cloud service providers organize the cloud computing environment as a multi-tier architecture. Each tier executes its designated tasks and passes the job to the next tier; in a fashion similar, but not identical, to the traditional job-shop environments. An optimization process must take place to schedule the appropriate tasks of the job on the resources of the tier, so as to meet the QoS expectations of the job. Existing approaches employ scheduling strategies that consider the performance optimization at the individual resource level and produce optimal single-tier driven schedules. Due to the sequential nature of the multi-tier environment, the impact of such schedules on the performance of other resources and tiers tend to be ignored, resulting in a less than optimal performance when measured at the multi-tier level.
    In this paper, we propose a multi-tier-oriented job scheduling and allocation technique. The scheduling and allocation process is formulated as a problem of assigning jobs to the resource queues of the cloud computing environment, where each resource of the environment employs a queue to hold the jobs assigned to it. The scheduling problem is NP-hard, as such a biologically inspired genetic algorithm is proposed. The computing resources across all tiers of the environment are virtualized in one resource by means of a single queue virtualization. A chromosome that mimics the sequencing and allocation of the tasks in the proposed virtual queue is proposed. System performance is optimized at this chromosome level. Chromosome manipulation rules are enforced to ensure task dependencies are met. The paper reports experimental results to demonstrate the performance of the proposed technique under various conditions and in comparison with other commonly used techniques.
    KEYWORDS

    Cloud Computing, Task Scheduling and Allocation, QoS Optimization, Load Balancing, Genetic Algorithms


    Virtual Enterprise Architecture Supply Chain (VEASC) Model On Cloud Computing: A Simulation-Based Study Through OPNET Modelling
    Tlamelo Phetlhu, Department of Commerce and Law, University of Zululand, KwaZulu Natal, South Africa
    ABSTRACT
    The virtual enterprise architecture supply chain (VEASC) model has been studied in this research employing OPNET modelling and simulations. VEASC requires a synchronous framework of integrated applications and databases, coordination, collaborations, and communications for ensuring high accuracy and responsiveness in a supply chain (SC). The traditional models of electronic data interchange (EDI) and out-of-application methods for messaging and collaborations are not suitable to achieve the full benefits of VEASC because multiple human interventions may be required. In this research, a cloud-based SC application and its distributed databases contributed by multiple supplier and buyer organisations are modelled and simulated on OPNET. The application modelled on the cloud is based on a commercial software called INTEND. The simulation results revealed continuous flow of all the phases of the SC application because the reports reflected continuous interactions between the agents involved and the cloud distributed databases. This model is a good enabler of the VEASC model.
    KEYWORDS

    Supply chain, enterprise architecture, integration, collaboration, cloud computing, OPNET, simulations


    SECURITY ISSUES IN CLOUD-BASED BUSINESSES
    Mohamad Ibrahim AL Ladan, Rafik Hariri University, LEBANON
    ABSTRACT
    Cloud-based Business is a Business running and relying on Cloud computing IT paradigm. Cloud computing is an emerging technology paradigm that transfers current technological and computing concepts into utility-like solutions similar to electricity and communication systems. It provides the full scalability, reliability, computing resources configurability and outsourcing, resource sharing, external data warehousing, and high performance and relatively low cost feasible solutions and services as compared to dedicated infrastructures. Cloud-based Businesses store, access, use, and manage their data and software applications over the internet on a set of servers in the cloud without the need to have them stored/installed locally on their local devices. The cloud technology is used daily by many businesses/people around the world from using web based email services to executing heavy complex business transactions. Like any other emerging technology, Cloud computing comes with a baggage of some pros and cons. It is very useful in business development as it brings amazing results in a timely manner; however, it comes with increasing security and privacy concerns and issues. In this paper we will investigate, analyse, classify, and discuss the new security concerns and issues introduced by cloud computing. In addition, we present some security requirements that address and may alleviate these concerns and issues.
    KEYWORDS

    Cloud-based Business security issues and concerns; Cloud computing security issues and concerns. Cloud computing security requirements


    A MapReduce based Algorithm for Data Migration in a Private Cloud Environment
    Anurag Kumar Pandey, Ruppa K. Thulasiram and A. Thavaneswaran, University of Manitoba, Winnipeg, Canada
    ABSTRACT
    Use of cloud computing has grown quickly and is now a technology that serves the compute, storage and data management needs of individuals, businesses, governments and other organizations. Large data centers are created to serve large number of clients' workload from various walks of life. When a resource in a data center reaches its end-of-life, instead of investing in upgrading, it is possibly the time to decommission such a resource and migrate workloads to other resources in the data center. Data migration between different cloud servers of a given private cloud is risky due to the possibility of data loss. We have designed a MapReduce based algorithm and have introduced few metrics to test and evaluate our proposed framework. We show that our algorithm for data migration works efficiently for text, image, audio and video files with minimum data loss and scale well for large data as well.
    KEYWORDS

    Cloud Computing, Private Cloud, Data Migration, MapReduce, Data Loss, Cost