Autonomic QoS control in enterprise Grid environments using online simulation

As Grid Computing increasingly enters the commercial domain, performance and quality of service (QoS) issues are becoming a major concern. The inherent complexity, heterogeneity and dynamics of Grid computing environments pose some challenges in managing their capacity to ensure that QoS requirements are continuously met. In this paper, a comprehensive framework for autonomic QoS control in enterprise Grid environments using online simulation is proposed. This paper presents a novel methodology for designing autonomic QoS-aware resource managers that have the capability to predict the performance of the Grid components they manage and allocate resources in such a way that service level agreements are honored. Support for advanced features such as autonomic workload characterization on-the-fly, dynamic deployment of Grid servers on demand, as well as dynamic system reconfiguration after a server failure is provided. The goal is to make the Grid middleware self-configurable and adaptable to changes in the system environment and workload. The approach is subjected to an extensive experimental evaluation in the context of a real-world Grid environment and its effectiveness, practicality and performance are demonstrated.     

Towards an autonomic performance management approach for a cloud broker environment using a decomposition–coordination based methodology

Efficient resource allocation of computational resources to services is one of the predominant challenges in a cloud computing environment. Furthermore, the advent of cloud brokerage and federated cloud computing systems increases the complexity of cloud resource management. Cloud brokers are considered third party organizations that work as intermediaries between the service providers and the cloud providers. Cloud brokers rent different types of cloud resources from a number of cloud providers and sublet these resources to the requesting service providers. In this paper, an autonomic performance management approach is introduced that provides dynamic resource allocation capabilities for deploying a set of services over a federated cloud computing infrastructure by considering the availability as well as the demand of the cloud computing resources. A distributed control based approach is used for providing autonomic computing features to the proposed framework via a feedback-based control loop. This distributed control based approach is developed using one of the decomposition–coordination methodologies, named interaction balance, for interactive bidding of cloud computing resources. The primary goals of the proposed approach are to maintain the service level agreements, maximize the profit, and minimize the operating cost for the service providers and the cloud broker. The application of interaction balance methodology and prioritization of profit maximization for the cloud broker and the service providers during resource allocation are novel contributions of the proposed approach.     

Coordinated load management in Peer-to-Peer coupled federated grid systems

This paper proposes a coordinated load management protocol for Peer-to-Peer?(P2P) coupled federated Grid systems. The participants in the system, such as the resource providers and the consumers who belong to multiple control domains, work together to enable a coordinated federation. The coordinated load management protocol embeds a logical spatial index over a Distributed Hash Table?(DHT) space for efficient management of the coordination objects; the DHT-based space serves as a kind of decentralized blackboard system. We show that our coordination protocol has a message complexity that is logarithmic to the number of nodes in the system, which is significantly better than existing broadcast based coordination protocols. The proposed load management protocol can be applied for efficiently coordinating resource brokering services of distributed computing systems such as grids and PlanetLab. Resource brokering services are the main components that control the way applications are scheduled, managed and allocated in a distributed, heterogeneous, and dynamic Grid computing environments. Existing Grid resource brokers, e-Science application work-flow schedulers, operate in tandem but still lack a coordination mechanism that can lead to efficient application schedules across distributed resources. Further, lack of coordination exacerbates the utilization of various resources (such as computing cycles and network bandwidth). The feasibility of the proposed coordinated load management protocol is studied through extensive simulations.     

Ontological Semantics for Distributing Contextual Knowledge in Highly Distributed Autonomic Systems

Much recent research has focused on applying Autonomic Computing principles to achieve constrained self-management in adaptive systems, through self-monitoring and analysis, strategy planning, and self adjustment. However, in a highly distributed system, just monitoring current operation and context is a complex and largely unsolved problem domain. This difficulty is particularly evident in the areas of network management, pervasive computing, and autonomic communications. This paper presents a model for the filtered dissemination of semantically enriched knowledge over a large loosely coupled network of distributed heterogeneous autonomic agents, removing the need to bind explicitly to all of the potential sources of that knowledge. This paper presents an implementation of such a knowledge delivery service, which enables the efficient routing of distributed heterogeneous knowledge to, and only to, nodes that have expressed an interest in that knowledge. This gathered knowledge can then be used as the operational or context information needed to analyze to the system's behavior as part of an autonomic control loop. As a case study this paper focuses on contextual knowledge distribution for autonomic network management. A comparative evaluation of the performance of the knowledge delivery service is also provided. John Keeney holds a BAI degree in Computer Engineering and a PhD in Computer Science from Trinity College Dublin. His primary interests are in controlling autonomic adaptable systems, particularly when those systems are distributed. David Lewis graduated in Electronics Engineering from the University of Southampton and gained his PhD in Computer Science from University College London. His areas of interest include integrated network and service management, distributed system engineering, adaptive and autonomic systems, semantic services and pervasive computing. Declan O’Sullivan was awarded his primary degree, MSc and PhD in Computer Science from Trinity College Dublin. He has a particular interest in the issues of semantic interoperability and heterogeneous information querying within a range of areas, primarily network and service management, autonomic management, and pervasive computing.     

ARMCO: Advanced topics in resource management for ubiquitous cloud computing: An adaptive approach

Cloud Computing can be seen as one of the latest major evolution in computing offering unlimited possibility to use ICT in various domains: business, smart cities, medicine, environmental computing, mobile systems, design and implementation of cyber-infrastructures. The recent expansion of Cloud Systems has led to adapting resource management solutions for large number of wide distributed and heterogeneous datacenters. The adaptive methods used in this context are oriented on: self-stabilizing, self-organizing and autonomic systems; dynamic, adaptive and machine learning based distributed algorithms; fault tolerance, reliability, availability of distributed systems. The pay-per-use economic model of Cloud Computing comes with a new challenge: maximizing the profit for service providers, minimizing the total cost for customers and being friendly with the environment.This special issue presents advances in virtual machine assignment and placement, multi-objective and multi-constraints job scheduling, resource management in federated Clouds and in heterogeneous environments, dynamic topology for data distribution, workflow performance improvement, energy efficiency techniques and assurance of Service Level Agreements.     

一种适应数据与计算密集型任务的私有云系统实现研究*

与公有云计算相比,针对数据与计算双重密集型任务的私有云计算系统对计算效率和系统管理效率提出了更高的要求,目前的公有云计算系统显得过于复杂和繁琐,因此需要一种简便易用的能够适应数据与计算密集型任务的私有云计算系统实现。借鉴公有云计算的相关理论和实现方法,提出了一种针对数据与计算双重密集型任务的私有云计算系统实现方案。该方案通过作业文件描述用户的计算任务,确定计算任务的计算模型和计算的输入输出文件;针对私有云的特点,简化Google云计算系统的MapReduce并行处理框架,得到更加直观的数据计算模型;自动连     

Attribute-based data access control in mobile cloud computing: Taxonomy and open issues

With the thriving growth of the cloud computing, the security and privacy concerns of outsourcing data have been increasing dramatically. However, because of delegating the management of data to an untrusted cloud server in data outsourcing process, the data access control has been recognized as a challenging issue in cloud storage systems. One of the preeminent technologies to control data access in cloud computing is Attribute-based Encryption (ABE) as a cryptographic primitive, which establishes the decryption ability on the basis of a user’s attributes. This paper provides a comprehensive survey on attribute-based access control schemes and compares each scheme’s functionality and characteristic. We also present a thematic taxonomy of attribute-based approaches based on significant parameters, such as access control mode, architecture, revocation mode, revocation method, revocation issue, and revocation controller. The paper reviews the state-of-the-art ABE methods and categorizes them into three main classes, such as centralized, decentralized, and hierarchal, based on their architectures. We also analyzed the different ABE techniques to ascertain the advantages and disadvantages, the significance and requirements, and identifies the research gaps. Finally, the paper presents open issues and challenges for further investigations.     

Denial of service attacks and defenses in decentralized trust management

Trust management is an approach to scalable and flexible access control in decentralized systems. In trust management, a server often needs to evaluate a chain of credentials submitted by a client; this requires the server to perform multiple expensive digital signature verifications. In this paper, we study low-bandwidth Denial-of-Service (DoS) attacks that exploit the existence of trust management systems to deplete server resources. Although the threat of DoS attacks has been studied for some application-level protocols such as authentication protocols, we show that it is especially destructive for trust management systems. Exploiting the delegation feature in trust management languages, an attacker can forge a long credential chain to force a server to consume a large amount of computing resource. Using game theory as an analytic tool, we demonstrate that unprotected trust management servers will easily fall prey to a witty attacker who moves smartly. We report our empirical study of existing trust management systems, which manifests the gravity of this threat. We also propose a defense technique using credential caching, and show that it is effective in the presence of intelligent attackers. A preliminary version of this paper was presented at the Second IEEE International Conference on Security and Privacy in Communication Networks, Baltimore, MD, USA, August 2006.     

Fine-grained management of I/O optimizations based on workload characteristics

With the advent of new computing paradigms, parallel file systems serve not only traditional scientific computing applications but also non-scientific computing applications, such as financial computing, business, and public administration. Parallel file systems provide storage services for multiple applications. As a result, various requirements need to be met. However, parallel file systems usually provide a unified storage solution, which cannot meet specific application needs. In this paper, an extended file handle scheme is proposed to deal with this problem. The original file handle is extended to record I/O optimization information, which allows file systems to specify optimizations for a file or directory based on workload characteristics. Therefore, fine-grained management of I/O optimizations can be achieved. On the basis of the extended file handle scheme, data prefetching and small file optimization mechanisms are proposed for parallel file systems. The experimental results show that the proposed approach improves the aggregate throughput of the overall system by up to 189.75%.     

基于MEC的蜂窝网络联合计算与无线资源管理

移动边缘计算（MEC）是当下最有效的增强移动设备计算能力的方法，吸引了广大学者进行研究。为提高移动边缘计算的无线蜂窝网络性能，提出了一种基于移动边缘计算的无线蜂窝网络计算卸载和干扰管理集成框架。在该集成框架中，MEC服务器综合基于所有用户设备（UE）估算的计算开销和由MEC服务器自身估算的卸载开销做出卸载决策。然后MEC服务器再使用图着色进行PRB分配。最后基于卸载决策和PRB分配结果将MEC服务器的计算资源分配给用户设备UE。对该集成框架的仿真结果展现了该集成框架在不同系统参数下的有效性。     

ANEMA: Autonomic network management architecture to support self-configuration and self-optimization in IP networks

The increasing scale, technology advances and services of modern networks have dramatically complicated their management such that in the near future it will be almost impossible for human administrators to monitor them. To control this complexity, IBM has introduced a promising approach aiming to create self-managed systems. This approach, called Autonomic Computing, aims to design computing equipment able to self-adapt its configuration and to self-optimize its performance depending on its situation in order to fulfill high-level objectives defined by the human operator.In this paper, we present our autonomic network management architecture (ANEMA) that implements several policy forms to achieve autonomic behaviors in the network equipments. In ANEMA, the high-level objectives of the human administrators and the users are captured and expressed in terms of ‘Utility Function’ policies. The ‘Goal’ policies describe the high-level management directives needed to guide the network to achieve the previous utility functions. Finally, the ‘behavioral’ policies describe the behaviors that should be followed by network equipments to react to changes in their context and to achieve the given ‘Goal’ policies.In order to highlight the benefits of ANEMA architecture and the continuum of policies to introduce autonomic management in a multiservice IP network, a testbed has been implemented and several scenarios have been executed.     

基于状态的通用自主计算模型

针对目前由于计算系统庞大造成管理工作复杂以及处理突发事件低效的问题,提出一种通用的自主计算模型.通过归纳系统的运行状态和状态变化,利用统计记分方法对状态变化进行判定和标记,实现了包含自我配置、恢复、优化和保护在内的自主计算特性,使系统可以自主地规划最能够满足系统运行需求的方案,从而将繁重的系统配置和突发事件处理任务由管理人员移交给系统自身.实验结果表明,该运用该模型能够在减轻管理负担的情况下,自主地实现系统资源的合理配置与使用.     

自主计算软件工程方法

从软件工程角度对自主计算进行研究，定义了一种自、主计算系统模型．以此模型为基础，提出了一种多主体系统设计方法DPMAS用于自主计算系统的建模，提出了一种基于构件复用的层次化自主计算未统开发方法．根据研究成果，实现了一个自主计算开发环境，较好的支撑了自主计算系统的软件工程过程．     

Cloud-Sea Computing Systems: Towards Thousand-Fold Improvement in Performance per Watt for the Coming Zettabyte Era

We are entering a new era of computing, characterized by the need to handle over one zettabyte （1021 bytes, or ZB） of data. The world＇s capacities to sense, transmit, store, and process information need to grow three orders of magnitude, while maintain an energy consumption level similar to that of the year 2010. In other words, we need to produce thousand-fold improvement in performance per watt. To face this challenge, in 2012 the Chinese Academy of Sciences launched a 10-year strategic priority research initiative called the Next Generation Information and Communication Technology initiative （the NICT initiative）. A research thrust of the NICT program is the Cloud-Sea Computing Systems project. The main idea is to augment conventional cloud computing by cooperation and integration of the cloud-side systems and the sea-side systems, where the ＂sea-side＂ refers to an augmented client side consisting of human facing and physical world facing devices and subsystems. The Cloud-Sea Computing Systems project consists of four research tasks： a new computing model called REST 2.0 which extends the REST （representational state transfer） architectural style of Web computing to cloud-sea computing, a three-tier storage system architecture capable of managing ZB of data, a billion-thread datacenter server with high energy efficiency, and an elastic processor aiming at energy efficiency of one trillion operations per second per watt. This special section contains 12 papers produced by the Cloud-Sea Computing Systems project team, presenting research results relating to sensing and REST 2.0, the elastic processor, the hyperparallel server, and the cloud-sea storage.     

自主计算概念模型与实现方法

自主计算是一个新兴的热点研究领域,旨在通过"技术管理技术"的手段隐藏系统管理复杂性,建立用户可指导的、状态觉察的和自适应的计算机系统.目前,自主计算的研究仍处于起步阶段,尚缺乏系统而成熟的理论体系.在阐明自主计算概念的基础上,提出一个自主计算概念模型.该模型刻画了自主元素和自主计算系统的基本工作机制和原理.以该模型为依据,概括性地提出了两类分别基于知识模型和数学模型的自主计算系统,分析它们的优点和不足.最后,给出了自主计算研究展望.     

An empirical study of modeling self-management capabilities in autonomic systems using case-based reasoning

Autonomic systems promise to inject self-managing capabilities in software systems. The major objectives of autonomic computing are to minimize human intervention and to enable a seamless self-adaptive behavior in the software systems. To achieve self-managing behavior, various methods have been exploited in past. Case-based reasoning (CBR) is a problem solving paradigm of artificial intelligence which exploits past experience, stored in the form of problem–solution pairs. We have applied CBR based modeling approach to achieve autonomicity in software systems. The proposed algorithms have been described and CBR implementation on externalization and internalization architectures of autonomic systems using two case studies RUBiS and Autonomic Forest Fire Application (AFFA) have been shown. The study highlights the effect of 10 different similarity measures, the role of adaptation and the effect of changing nearest neighborhood cardinality for a CBR solution cycle in autonomic managers. The results presented in this paper show that the proposed CBR based autonomic model exhibits 90–98% accuracy in diagnosing the problem and planning the solution.     

DTM:一种面向网络计算的动态信任管理模型

在开放的互联网中,信任关系的建立是实现跨自治域资源共享与协同的前提.然而,网络计算环境的分布自治特性,使得各类复杂系统中应用不同的信任管理机制,容易导致信任定义的不一致问题,特别是很多系统为支持多域协作,直接假设实体间信任具有传递特性,而在模型中缺乏该性质成立的条件;此外,网络计算环境中的动态演化特性,使得驱动实体间协作的信任关系随需而变,而现有信任管理模型仅仅关注于系统功能结构,缺乏对这种动态性的描述.文中提出了一种动态信任管理模型DTM,基于信念公式形式化定义了主体间的信任公式,并将信任的传递特性(信任链)解释为模型的一条性质.在该模型中,针对信任关系的动态特征,以时间为参量刻画主体公式集,以事件为触发条件刻画主体间信任的变化,并基于正则事件序列描述信任管理的资源授权过程,可刻画主体间信任的建立过程.最终,设计、实现了一个信任管理系统CROWN-TM,并进行了初步实验分析.     

大数据管理系统的历史、现状与未来

大数据管理技术正在经历以软件为中心到以数据为中心的计算平台的变迁,传统的关系型数据库管理系统无法满足现在以数据为中心的大数据管理的需求,设计新型大数据管理系统迫在眉睫.首先回顾了数据管理技术的发展历史;之后,从大数据管理的存储、数据模型、计算模式、查询引擎等方面分析了大数据管理系统的现状,指出目前大数据管理系统具有模块化和松耦合的特点,并进一步介绍了大数据管理系统应具备的数据特征、系统特征和应用特征,指出大数据管理系统技术还在快速进化之中,预测未来的大数据管理系统应具备多数据模型并存、多计算模式融合、可伸缩调整、新硬件驱动、自适应调优等特点.     

A Middleware-based Distributed Management Framework for Next-generation Network

As the basis for next-generation application middleware, distributed computing technique is increasingly used in the network management to fulfill the integration of diverse management systems. Furthermore Web technology is widely used in interaction friendly with user. This article puts forward a MNMI (Middleware-based Network Management Integration) framework in the domain of the management of IPv6 test bed-NSFCNET (National Natural Science Foundation of China Network). The framework uses a logically-layers method from user interface to managers and pseudo-agent, which uses distributed management and groups managed objects to retrieve, and thus it has assured the performance of network management by providing multiple object quick access.     

An empirical study of modeling self-management capabilities in autonomic systems using case-based reasoning

Autonomic systems promise to inject self-managing capabilities in software systems. The major objectives of autonomic computing are to minimize human intervention and to enable a seamless self-adaptive behavior in the software systems. To achieve self-managing behavior, various methods have been exploited in past. Case-based reasoning (CBR) is a problem solving paradigm of artificial intelligence which exploits past experience, stored in the form of problem–solution pairs. We have applied CBR based modeling approach to achieve autonomicity in software systems. The proposed algorithms have been described and CBR implementation on externalization and internalization architectures of autonomic systems using two case studies RUBiS and Autonomic Forest Fire Application (AFFA) have been shown. The study highlights the effect of 10 different similarity measures, the role of adaptation and the effect of changing nearest neighborhood cardinality for a CBR solution cycle in autonomic managers. The results presented in this paper show that the proposed CBR based autonomic model exhibits 90–98% accuracy in diagnosing the problem and planning the solution.