Computing and Information Systems - Theses
Permanent URI for this collection
Search Results
1 - 4 of 4
-
ItemCost-efficient resource provisioning for large-scale graph processing systems in cloud computing environments( 2018)A large amount of data that is being generated on Internet every day is in the form of graphs. Many services and applications namely as social networks, Internet of Things (IoT), mobile applications, business applications, etc. in which every data entity can be considered as a vertex and the relationships between entities shape the edges of a graph, are in this category. Since 2010, exclusive large-scale graph processing frameworks are being developed to overcome the inefficiency of traditional processing solutions such as MapReduce. However, most frameworks are designed to be employed on high performance computing (HPC) clusters which are only available to whom can afford such infrastructure. Cloud computing is a new computing paradigm that offers unprecedented features such as scalability, elasticity and pay-as-you-go billing model and is accessible to everyone. Nevertheless, the advantages that cloud computing can bring to the architecture of large-scale graph processing systems are less studied. Resource provisioning and management is a critical part of any processing system in cloud environments. To provide the optimized amount of resources for a particular operation, several factors such as monetary cost, throughput, scalability, network performance, etc. can be taken into consideration. In this thesis, we investigate and propose novel solutions and algorithms for cost-efficient resource provisioning for large-scale graph processing systems. The outcome is a series of research works that increase the performance of such processing by making it aware of the operating environment while decreasing the dollar cost significantly. We have particularly made the following contributions: 1. We introduced iGiraph, a cost-efficient framework for processing large-scale graphs on public clouds. iGiraph also provides a new graph algorithm categorization and processes the graph accordingly. 2. To demonstrate the impact of network on the processing in cloud environment, we developed two network-aware algorithms that utilize network factors such as traffic, bandwidth and also the computation power. 3. We developed an auto-scaling technique to take advantage of resource heterogeneity on clouds. 4. We introduced a large-scale graph processing service for clouds where we consider the service level agreement (SLA) requirements in the operations. The service can handle multiple processing requests by its new prioritization and provisioning approach.
-
ItemIntegrated provisioning of compute and network resources in Software-Defined Cloud Data Centers( 2018)Software-Defined Networking (SDN) has opened up new opportunities in networking technology with its decoupled concept of the control plane from the packet forwarding hardware, which enabled the network to be programmable and configurable dynamically through the centralized controller. Cloud computing has been empowered with the adoption of SDN for infrastructure management in a data center where dynamic controllability is indispensable in order to provide elastic services. The integrated provisioning of compute and network resources enabled by SDN is essential in clouds to enforce reasonable Service Level Agreements (SLAs) stating the Quality of Service (QoS) while saving energy consumption and resource wastage. This thesis presents the joint compute and network resource provisioning in SDN-enabled cloud data center for QoS fulfillment and energy efficiency. It focuses on the techniques for allocating virtual machines and networks on physical hosts and switches considering SLA, QoS, and energy efficiency aspects. The thesis advances the state-of-the-art with the following key contributions: 1. A taxonomy and survey of the current research on SDN-enabled cloud computing, including the state-of-the-art joint resource provisioning methods and system architectures. 2. A modeling and simulation environment for SDN-enabled cloud data centers abstracting functionalities and behaviors of virtual and physical resources. 3. A novel dynamic overbooking algorithm for energy efficiency and SLA enforcement with the migration of virtual machines and network flows. 4. A QoS-aware computing and networking resource allocation algorithm based on the application priority to fulfill different QoS requirements. 5. A prototype system of the integrated control platform for joint management of cloud and network resources simultaneously based on OpenStack and OpenDaylight.
-
ItemResource provisioning and scheduling algorithms for scientific workflows in cloud computing environments( 2016)Scientific workflows describe a series of computations that enable the analysis of data in a structured and distributed manner. Their importance is exacerbated in todays big data era as they become a compelling mean to process and extract knowledge from the ever-growing data produced by increasingly powerful tools such as telescopes, particle accelerators, and gravitational wave detectors. Due to their large-scale nature, scheduling algorithms are key to efficiently automate their execution in distributed environments, and as a result, to facilitate and accelerate the pace of scientific progress. The emergence of the latest distributed system paradigm, cloud computing, brings with it tremendous opportunities to run workflows at low costs without the need of owning any infrastructure. In particular, Infrastructure as a Service (IaaS) clouds, offer an easily accessible, flexible, and scalable infrastructure for the deployment of these scientific applications by providing access to a virtually infinite pool of resources that can be acquired, configured, and used as needed and are charged on a pay-per-use basis. This thesis investigates novel resource provisioning and scheduling approaches for scientific workflows in IaaS clouds. They address fundamental challenges that arise from the multi-tenant, resource-abundant, and elastic resource model and are capable of fulfilling a set of quality of service requirements expressed in terms of execution time and cost. It advances the field by making the following key contributions: 1. A taxonomy and survey of the state-of-the-art scientific workflow scheduling algorithms designed exclusively for IaaS clouds. 2. A novel static scheduling algorithm that leverages Particle Swarm Optimization to generate a workflow execution and resource provisioning plan that minimizes the infrastructure cost while meeting a deadline constraint. 3. A hybrid algorithm based on a variation of the Unbounded Knapsack Problem that finds a trade-off between making static decisions to find better-quality schedules and dynamic decisions to adapt to unexpected delays. 4. A scalable algorithm that combines heuristics and two different Integer Programming models to generate schedules that minimize the execution time of the work- flow while meeting a budget constraint. 5. The implementation of a cloud resource management module and its integration to an existing Workflow Management System.
-
ItemSLA-based resource provisioning for management of Cloud-based Software-as-a-Service applications( 2014)The Cloud computing Software-as-a-Service (SaaS) model has changed the sales model for software providers. The SaaS model transforms the traditional license based model to a subscription model, which allows customers to access applications over the Internet without software and hardware upfront costs and provides reduced maintenance costs. However, the key for sales is still customer satisfaction which is at the heart of the selling process. To guarantee Quality of Service (QoS) for customer satisfaction therefore, the Service Level Agreement (SLA) is implemented between customers and SaaS providers, where the main objectives are profit maximization and increased market share. To achieve these objectives, there are several challenges due to the dynamic nature of the Cloud environment. Firstly, the SaaS provider utilizes shared infrastructure and various types of request loads which can lead to unpredictability in performance and availability of resources. Secondly, there is a possibility that existing customers may make changes in requirements, which can lead to resource reallocation. As such, resource allocation may cause SLA violations which could reduce the SaaS providers’ profit margin and reputation, meaning a possible loss of existing customers and potential new customers. Thirdly, SaaS providers need to attract customers with special needs and consider market competition from other providers in order to increase profit and market share. To overcome the above challenges, most proposed solutions are focused on the resource management with the aim of minimizing cost without sufficient consideration of customer needs. Therefore, to address these challenges, this thesis proposes algorithms and techniques for optimal provisioning of Cloud resources with the aim of maximizing profit and customer base by handling the dynamism associated with SLAs and heterogeneous resources. The key contributions of the thesis are: • A comprehensive survey of how SLAs are created, managed and used with case examples drawn from both academy and industry with a major emphasis on the SLA-based resource management systems. • The admission control and scheduling algorithms assist in identifying which request is more acceptable based on profitability, reducing the probability of SLA violations given the heterogeneous nature of Cloud resources. • The customer requirements driven resource provisioning algorithms can help in adapting to changes in the requirements. The proposed algorithms provide personalized attention to the customer and are also able to understand specific customer needs. • A new negotiation framework to enlarge a SaaS provider’s customer base that considers dynamism in the Cloud environment with time and market factors to make the best possible decisions for negotiation. • A prototype of the customer requirements driven SLA-based resource management system to prove the usefulness of our proposed strategies using the latest technologies.