Skip to main content

2013 | Buch

Dynamic Cloud Collaboration Platform

A Market-Oriented Approach

insite
SUCHEN

Über dieses Buch

Present trends in cloud providers (CPs) capabilities have given rise to the interest in federating or collaborating clouds, thus allowing providers to revel on an increased scale and reach more than that is achievable individually. Current research efforts in this context mainly focus on building supply chain collaboration (SCC) models, in which CPs leverage cloud services from other CPs for seamless provisioning. Nevertheless, in the near future, we can expect that hundreds of CPs will compete to offer services and thousands of users will also compete to receive the services to run their complex heterogeneous applications on a cloud computing environment. In this open federation scenario, existing collaboration models (i.e. SCC) are not applicable since they are designed for static environments where a-priori agreements among the parties are needed to establish the federation.

To move beyond these shortcomings, Dynamic Cloud Collaboration Platform establishes the basis for developing dynamic, advanced and efficient collaborative cloud service solutions that are scalable, high performance, and cost effective. We term the technology for inter-connection and inter-operation of CPs in open cloud federation as Dynamic Cloud Collaboration (DCC), in which various CPs (small, medium, and large) of complementary service requirements will collaborate dynamically to gain economies of scale and enlargements of their capabilities to meet quality of service (QoS) requirements of consumers. In this context, this book addresses four key issues - when to collaborate (triggering circumstances), whom to collaborate with (suitable partners), how to collaborate (architectural model), and how to demonstrate collaboration applicability (simulation study). It also provides solutions, which are effective in real environments.

Inhaltsverzeichnis

Frontmatter
Chapter 1. Overview of Cloud Computing and Motivation of the Work
Abstract
This chapter provides an overview of cloud computing, which can help the readers to understand the basics of cloud computing technology as well as motivation of the current work towards enabling dynamic cloud federation platform.
Mohammad Mehedi Hassan, Eui-Nam Huh
Chapter 2. Related Work
Abstract
In this chapter, first, we provide an overview of present cloud collaboration or federation initiatives. Second, we discuss existing works on market-oriented cloud models. Finally, existing collaborator or partner selection approaches in different areas are presented.
Mohammad Mehedi Hassan, Eui-Nam Huh
Chapter 3. Architectural Framework and Market Model for Dynamic Cloud Collaboration
Abstract
Existing cloud providers, operating in isolation, are often prone to Service Level Agreement violations and resources over-provisioning in order to ensure high-quality services to end-users, thus incurring extensive operational cost and labor. As mentioned in Chap. 1, dynamic cloud collaboration (DCC) is an approach to reduce expenses and avoid adverse business impact. It is formed by a set of autonomous cloud providers who cooperate through a mechanism to share resources while enjoying larger scale and reach. This chapter first presents the architecture that establishes the basis to form DCC. Finally, it describes the proposed combinatorial auction (CA)-based cloud market model called CACM that enables and commercializes a DCC platform.
Mohammad Mehedi Hassan, Eui-Nam Huh
Chapter 4. Multi-objective Optimization Model and Algorithms for Partner Selection
Abstract
The Partner selection is an important decision problem in the formation of a dynamic cloud collaboration platform. Selecting suitable cloud partners to form a group will facilitate the success of collaborative cloud services. In this chapter, first, we present a promising multi-objective (MO) optimization model of partner selection considering individual information (INI) and past relationship information (PRI) with collaboration cost optimization among cloud providers in a DCC platform. Then to solve this MO optimization model, a general framework of multi-objective genetic algorithm (MOGA) that uses INI and PRI of cloud providers called MOGA-IC is presented. Finally, two algorithms called NSGA-II and SPEA2 are developed to implement MOGA-IC.
Mohammad Mehedi Hassan, Eui-Nam Huh
Chapter 5. Experimental Results and Analysis
Abstract
In this chapter, we present our evaluation methodology and simulation results of the utility of dynamic cloud collaboration (DCC), MOGA-IC for CP partner selection and proposed CACM model. First, we compare DCC platform with existing cloud federation model. Then, we present the comparison of the proposed CACM model with the existing CA model in terms of economic efficiency. Next, we present a simulation example of a partner selection problem (PSP) for a pCP in the CACM model. It is used to illustrate the proposed MOGA-IC method. NSGA-II and SPEA2 are utilized to develop the MOGA-IC. Further simulation examples are conducted to pinpoint the most viable approach (NSGA-II or SPEA2) for MOGA-IC. Moreover, we implement the existing MOGA that uses only INI called MOGA-I for CP partner selection and analyze its performance with MOGA-IC in the proposed CACM model. Finally, numerical results are presented to demonstrate the utility of the price minimization algorithm among cloud providers in the CACM model.
Mohammad Mehedi Hassan, Eui-Nam Huh
Chapter 6. Closing Remarks
Abstract
In today’s world the emerging cloud computing offers a new computing model where resources such as computing power, storage, online applications and networking infrastructures can be shared as “services” over the Internet. However, the prevalent commercial CPs, operating in isolation (i.e. proprietary in nature), often face resource over-provisioning, degraded performance, and SLA violations (i.e. cloud service outages), thus incurring high operational costs and limiting the scope and scale of their services. Present trends in cloud service providers capabilities give rise to the interest in federating or collaborating clouds, hence allowing providers to revel on increased scale and reach than that is achievable individually.
Mohammad Mehedi Hassan, Eui-Nam Huh
Backmatter
Metadaten
Titel
Dynamic Cloud Collaboration Platform
verfasst von
Mohammad Mehedi Hassan
Eui-Nam Huh
Copyright-Jahr
2013
Verlag
Springer New York
Electronic ISBN
978-1-4614-5146-4
Print ISBN
978-1-4614-5145-7
DOI
https://doi.org/10.1007/978-1-4614-5146-4