基于领域划分的公共云信任优化模型

151 浏览量更新于2024-08-26 收藏 363KB PDF 举报

"基于域划分的不可靠公共云信任模型"是一项针对当前公共云环境中面临的信任问题的研究论文。随着云计算的快速发展，公共云已经成为科学计算和商业应用的重要平台，其背后承载着大量的计算资源和数据。然而，云服务的可靠性问题日益突出，包括恶意提供商带来的安全威胁和性能不佳导致的服务质量下降。传统的信任计算方法往往依赖于高开销的验证机制，这在一定程度上影响了云系统的整体性能。为了克服这些问题，本文提出了一种创新的信任模型，它通过引入信任证书权威机构来管理计算域信任和全局信任。该模型的核心理念是将信任管理分解到各个独立的计算域中，通过域划分的方式降低信任计算的复杂性。这种方法减少了对系统资源的需求，提高了效率。作者Pei Yun Zhang（IEEE会员）和Yang Kong来自安徽师范大学数学与计算机科学学院，而Meng Chu Zhou（IEEE Fellow）则在新泽西理工学院的电气与计算机工程系任职。他们采用滑动窗口技术实时更新信任值，这种方法可以动态反映云服务的实时表现，从而做出更为准确的信任评估。实验结果显示，所提出的信任模型有效地减少了计算负担，提高了云服务的可信度，并能够在面对不可靠的服务时提供有效的信任决策支持。这种改进的模型对于提升公共云的稳定性、可靠性和安全性具有重要意义，为云计算环境下的信任管理提供了新的理论依据和技术支撑。这篇论文深入探讨了在公共云环境中如何通过智能的领域划分和动态信任管理策略来应对不可靠的服务挑战，为云服务提供商和用户之间的信任建立提供了一个实用且高效的解决方案。这一研究对于推动云计算行业的健康发展有着积极的影响。"

A Novel Trust Model for Unreliable Public Clouds

based on Domain Partition

PeiYun Zhang, Member, IEEE, Yang Kong

School of Mathematics and Computer Science

Anhui Normal University

Wuhu, Anhui 241003, China.

zpyanu@ahnu.edu.cn and yangkong2012@gmail.com

MengChu Zhou, Fellow, IEEE

Department of Electrical and Computer Engineering

New Jersey Institute of Technology

Newark, NJ 07102, USA.

zhou@njit.edu

Abstract—

Cloud computing has become an important

scientific computing and commercial application paradigm.

There are many computing resources and data in public

clouds, but there exist some threats caused by unreliable

services due to malicious providers or unacceptably poor

service performance. Traditional trust computing requires a

high overhead and thus decreases the performance of a

cloud system. To address such issues, this work proposes a

trust model by adopting trust certificate authority to

compute domain trust and global trust. The model

decreases computational complexity based on domain

partition. Sliding-windows are used to obtain updated

trust

values. Experimental results show that the proposed trust

model can achieve efficient and accurate trust computation.

Keywords.

Cloud computing; trust certificate authority; domain

partition; trust

NTRODUCTION

Cloud computing has experienced more than 10 years of

research and development. It brings convenience and simplicity

to users and service providers, such as through federation of

cloud providers [1]. More and more enterprises reduce

computing-related cost through service outsourcing with

maturity cloud computing technology.

Because of highly dynamic, distributed nature of cloud

services, trust computing is considered as one of the vital

challenges for

many high-security-demanding cloud

applications. According to the researches at University of

California at Berkeley, trust management and security are

ranked among the top 10 obstacles for adopting cloud

computing [2]. Unreliable cloud services may result in leakage

of user privacy data or easy-to-launch attacks. For example, a

software glitch at Dropbox (a USA-based cloud-storage

provider) temporarily allowed visitors to log in to any of its 25

million customers’ accounts using any password or none at all a

few years ago. Amazon’s Elastic Compute Cloud service

crashed during a system upgrade, knocking customers’ websites

off-line ranging from several hours to several days [3].

Traditional methods such Service-Level Agreement (SLA)

are inadequate for the trust management of complex cloud [4].

Customers facing problems identify a trustworthy cloud

provider solely on the basis of its SLA [5]. Providing new

secure and reliable trust management approaches in cloud has

been one of important foci of current cloud computing research.

II.

WORKS

Since trust was introduced from sociology into the field of

computers in 1994, trust has brought new security mechanisms

to the computer field. Trust is important for service management

in cloud environments. Li et al. propose an adaptive and

attribute-based trust model [4]. Abdul-Rahman et al. propose

direct trust and indirect trust, and calculate trust through a model

and obtain trust transfer values among entities [6]. Shao et al.

propose a trust model for depicting indirect trust [7]. Kamvar et

al. put forward the EigenTrust model to compute global trust

based on local trust and transaction information of related nodes

[8], but the model does not differentiate direct trust and indirect

one, while the differentiation is very important for trust

computation. Xiong et al. present a PeerTrust model that uses

feedback trust values of nodes. The model has a strong ability to

resist malicious nodes, but convergence of their algorithm is

rather slow [9]. Santos et al. introduce trust into clouds to

improve security and reliability of cloud services [10]. Beth et al.

propose the concepts of trust relationship among entities and

compute trust transfer values by using a probability and

statistics-based method [11]. Yan et al. propose a trust transfer

model [12]. Park et al. build a secure trust computing model

based on a security protocol [13]. Xing et al. take trust into

cloud scheduling to improve reliability and successful rate of a

schedule [14]. Noor et al. propose a reputation-based trust

management framework that provides a set of functionalities to

deliver trust as a service [15]. The above models or methods

have problems such as large overhead or low performance.

Trust models can be divided into two categories depending

upon whether they rely on reliable third party or not. The

representative of the first category is based on public key

infrastructure (PKI) [16]. It can achieve strong transitive trust.

One or a group of reliable nodes issue trust certificates, store and

update trust values in real time. The trust model can compute

effective trust values, but it brings about great resource overhead.

The representative of the second category is based on a trust

下载后可阅读完整内容，剩余5页未读，立即下载

weixin_38743235

粉丝: 10
资源: 941

基于领域划分的公共云信任优化模型

基于凸函数证据理论的关联感知云服务信任模型.pdf

域分区技术：优化不可靠云信任模型

用于基于云计算的电信服务的域划分的可配置安全性模型

P2P环境下基于兴趣的分布式信任模型.pdf

云计算环境下基于评价可信度的动态信任评估模型

基于用户行为的移动通信网业务域信任模型研究

SDN多控制器部署：基于拓扑划分的可靠性增强策略

基于子空间划分的模糊系统模型辨识算法研究

多域信任下的P2P访问控制模型：有效性与仿真实验

系统模型与模型化：基于可达矩阵的区域划分

最新资源