Evolving Developments in Grid and Cloud Computing: Advancing Research

Evolving Developments in Grid and Cloud Computing: Advancing Research

Emmanuel Udoh (Sullivan University, USA)
Release Date: January, 2012|Copyright: © 2012 |Pages: 383
ISBN13: 9781466600560|ISBN10: 146660056X|EISBN13: 9781466600577|DOI: 10.4018/978-1-4666-0056-0

Description

Grid and cloud computing both facilitate an increase in computing resources by the development of new connections to existing systems.

Evolving Developments in Grid and Cloud Computing: Advancing Research contains investigations of grid and cloud evolution, workflow management, and the impact new computing systems have on education and industry. Targeted at both researchers and IT professionals, this book provides current trends and emerging issues in cloud and grid architectures, standards and performance analysis.

Topics Covered

The many academic areas covered in this publication include, but are not limited to:

  • Balanced Job Scheduling
  • Cloud Applications for Mobile Social Networking
  • Data Grids
  • Grid Computing Environments
  • Grid Data Streaming Applications
  • Large Scale P2P Networks
  • Network Architectures for Massively Multiplayer Online Games
  • Next generation networks
  • Peer-to-Peer Desktop Grids
  • Trusted and Scalable Grids

Reviews and Testimonials

The book highlights the direction of current research with sufficient materials to spur efforts in this field by government, industry, institutions and individuals. In that light, this book impacts the advance of cloud technology and global economy, and indeed the information age.

– Emmanuel UdohNational College, Fort Wayne, USA

Table of Contents and List of Contributors

Search this Book:
Reset

Preface

The discipline of computing is coming of age, and cloud computing embodies the maturity of this field with its clear path to software generation, transmission, distribution and control. With cloud-virtualized resources (platform, infrastructure, software), the different components of computing are unified into an easily manageable entity, as services will ultimately be provided from data centers over the Internet and on-demand. Shaping the cloud paradigm shift are contributions from varied fields like virtualization, service-oriented architecture, grid-utility computing and distributed systems. Certainly, cloud computing is still evolving, but the advent of Apple iCloud signals its mainstream acceptance.

As a major pillar in the growth of cloud computing, virtualization has been exploited to create the illusion of unlimited power for hardware platform, operating system, data, memory, and storage device or network resources. The consolidation of servers has enabled more to be done with less hardware, thus supporting more users per piece of hardware as well as delivering and running applications faster. In that vein, the cloud’s virtual resources are typically cheaper than dedicated physical resources connected to a personal computer or network. Virtualization, which is only one possible service that cloud can deliver, is not always necessary in cloud computing; however, it has helped to centralize administrative tasks while improving scalability and workloads.

In a similar development, service oriented architecture (SOA) is well suited for web development and applications that can run in the cloud (Internet) rather than on local hardware. It allows the connection of discrete chunks of information in real time with the power to orchestrate a new application by providing parts of the existing applications with a new interface that is event-driven. With SOA, programmers configure the services to run on the Internet in ways that allow combining and reusing data (integration of existing or legacy applications) in new applications. Instead of a full-scale development of a new program (from the scratch), SOA extends what exists to the cloud where live data is accessed and accessible.

Furthermore, the cloud is a natural evolution of grid and distributed computing with its IT-related capabilities and resources that are provided as services via the Internet and on-demand. The user accesses or consumes cloud resources without being required to amass detailed knowledge of the underlying technology. Grid computing has a goal, in which computer processing power is seen as a utility that clients can pay for only as needed. It harnesses unused computing resources by connecting different computers together. This gluing of computers creates enormous storage and CPU capacity for large-scale computing such as weather forecasting and predictive models in science. Cloud computing has improved the grid computing concepts by harnessing computer resources as a utility with on-demand resource provisioning. As a matter of fact, cloud combines grid computing with utility computing concepts (pay-per-use – metered or subscription approach), but strictly hides the grid middleware from the users. Thus cloud computing is the consolidation of many year of computing endeavors, that poignantly signals the maturity of the computing field, with the additional capacity to curb the menace of software piracy.

The new advances in cloud computing will greatly impact IT services resulting in improved computational and storage resources as well as service delivery. To keep educators, students, researchers and professionals abreast of advances in the cloud, grid and high performance computing, this book Evolving Developments in Grid and Cloud Computing: Advancing Research will provide coverage of topical issues in the discipline. It will shed light on concepts, protocols, applications, methods and tools in this emerging and disruptive technology.

This book is organized in five distinct sections, covering wide-ranging topics such as: (1) Introduction (2) Scheduling (3) Architecture (4) Security and (5) Applications.

Section 1. Introduction, provides an overview of cloud computing and how to harness it in the explosive growth of social media. Cloud computing and social media are in vogue. Individuals, communities and organizations actively use the increasingly popular social media as a means of communication and interactive dialogue with user-generated content. As an Internet-based system that is enabled by ubiquitously accessible and scalable communication techniques, cloud-computing efficiently harnesses information from social networking services (e.g. Facebook, Twitter and LinkedIn), context-based systems (location, activity, interests, etc.) and personal applications (call logs, contacts, email, and calendar) for multiple purposes. The chapter “Harnessing the Cloud for Mobile Social Networking Applications” by Juwel Rana and co-authors, focuses on agent-based system for semantic and semi-automated applications that utilize the cloud to enrich and simplify communication services, for instance by displaying information, prioritizing information, and dynamically managing groups of users. This system exploits the concept of aggregated social graphs, which are created from harnessed information about how people communicate. The paper harped as well on the challenges in achieving the envisioned system with demonstrable prototyping results.

Section 2. Scheduling, is an integral part of grid and cloud implementation. Although scheduling is complex, prioritizing and load balancing are crucial management processes in cloud environments. Efficient job-scheduling algorithms are necessary to assign jobs to resources in grids and clouds. Nikolaos Preve, in the chapter “Balanced Job Scheduling Based on Ant Algorithm for Grid Network”, proposed a new Ant Colony Optimization (ACO) algorithm for balanced job scheduling in the grid-enabled environment. To achieve this goal, the author showed how to balance the entire system load by minimizing the makespan of a given set of jobs. The experimental outcomes demonstrated the practicability and competitiveness of this approach compared with other job scheduling algorithms. In a similar vein, Geoffrey Falzon and Maozhen Li, provided a chapter on "Evaluating Heuristics for Scheduling Dependent Jobs in Grid Computing Environments." This work takes cognizance of the fact that scheduling plays a crucial role in mapping jobs to cloud and grid resources, especially challenging jobs that have dependencies, which can be represented as Direct Acyclic Graphs (DAGs). The article appraised twelve representative heuristics for dependent job scheduling under one set of common assumptions, recognizing that scheduling m jobs to n resources with an objective to achieve a minimum makespan can be NP-complete. To facilitate performance evaluation, a DAG simulator was implemented which provided a set of tools for DAG job configuration, execution, and monitoring. The outcomes of the DAG simulation are also presented in this book.

In another chapter, "Peer-to-Peer Desktop Grids Based on an Adaptive Decentralized Scheduling Mechanism," researchers Ali, Saleh, Sarhan and Azab proposed an adaptive fuzzy logic based mechanism for dynamic computing environment in which matchmaking is achieved between resource requirements of outstanding tasks and resource capabilities of available workers. The results of the feasibility studies show that in a real time environment the resource utilization of executing workers can be maximized without exceeding the maximum execution time of the task. The researchers concluded that increasing the number of subtasks could reduce the efficiency of FMA-based decentralized scheduling, especially in parallel execution. In another study, "Predictive File Replication on the Data Grids" - researchers Liao, Helian, Wu and Rashid proposed a decision-tree-based predictive file replication strategy that forecasts files’ future popularity based on the characteristics of the grids. As noticed in most replication methods, the popularity of files is monitored or could be complex functions calculated for the overall cost of whether or not a replication decision or a deletion decision should be issued. The proposed strategy has shown superb performance in terms of mean job time and effective network usage compared with the other two-replication strategies, LRU and Economic under OptorSim simulation environment.

Section 3. Architecture, examines a host of research directions beyond the practice of aiming at the enterprise as the endpoint of IT architectures. As a matter of necessity clouds have to accommodate differences in architecture requirements for various situations such as in healthcare or financial domains. In this sense, autonomic computing, in which systems manage themselves according to certain goals, is increasingly valued in cloud computing. Innovations in cloud computing especially in the areas of virtualization are demonstrating that the goals of autonomic computing can be realized to a practical degree, and that they could be useful in developing cloud architectures capable of sustaining and supporting varied systems. Currently, there are efforts to merge the cloud and grid architectures with service-oriented architecture, autonomic computing and other open standards platforms. A couple of chapters in this section detail these advances.

In order to tackle the challenges encountered in the trust asymmetry between users and the grid system, Yuhui Deng and Na Helian proposed architecture based on trusted computing. In the chapter titled “Single Attestation Image for a Trusted and Scalable Grid”, the researchers constructed a trusted grid as a flat ring that decentralizes the functionalities of Trusted Platform Module (TPM) across the ring. This architecture presents a single attestation image, which provides a transparent attestation to a scalable, large-scale, and dynamic trusted grid. The architecture also significantly reduces the frequency of attestations, thus alleviating the TPM bottleneck and improving the commercialization of grid and cloud technology. In another chapter, “Personal Storage Grid Architecture: Consuming Cloud Data Space Resources”, researchers Lim, Wu, Simon, Rashid and Helian, discussed the development of a modern file system in Linux that enables the consumption of cloud applications and the availability of the underlying disk space resource to the end-user. A web service was developed to support systems like online email accounts and online virtual disk space. A free online mail account was used to demonstrate the architecture and the IMAP protocol used for remote data spaces.

Furthermore, cloud architectures are being developed for systems like wireless sensors network (WSN), which are increasingly widely used. The current WSN architectures do not promote the sharing of data on an inter-WSN basis, hence the need for newer architecture. Researchers Patel and Pandey, in the paper “Design of SOA Based Framework for Collaborative Cloud Computing in Wireless Sensor Networks”, proposed the SOA-based architecture to support collaborating cloud computing in WSN. The architecture consists of layered service stack that has management, information, presentation and communication layers with all required services and repositories. Interactions between WSN, subscribers and other cloud are also presented as sequence diagrams. The proposed framework serves the cloud subscribers with wide range of queries on the data of multiple WSNs through suitable interface to solve large-scale problems.

In a similar development, the researchers Eftychiou, Vrusias and Antonopoulos to satisfy the growing need to manage and search the online information developed a distributed semantic-enabled architecture, which enforces semantic web technologies for resource discovery. The chapter - “A Semantic-Driven Adaptive Architecture for Large Scale P2P Networks”, describes the conceptual representation of network knowledge on the P2P system that is organized in two-layered super-peer architecture, which is shaped from the information provided by the nodes using collective intelligence methods. The focus here is the creation of a dynamic hierarchical semantic-driven P2P topology using the network’s collective intelligence. The unmanageable amounts of data are transformed into a repository of semantic knowledge, transforming the network into ontology of conceptually related entities of information collected from the resources located by peers. Appropriate experiments have been undertaken through a case study by simulating the proposed architecture and evaluating results.

Online gaming software is growing in sophistication to meet the demand of users. Ma and Oikonomou reviewed various approaches on data replication and region partitioning used in some online games such as Eve Online, World of Warcraft and Second Life. These games are built on distributed client-server architectures with server allocation based on sharding, static geographical partitioning, dynamic micro-cell scheme, or optimal server. In the chapter titled “Network Architectures and Data Management for Massively Multiplayer Online Games”, these games are described as Current-generation Massively Multiplayer Online Games (MMOG) that allows players to be placed on a virtual region according to the geographical dispersion of players. The researchers investigated data storage and synchronization methods for MMOG databases, mainly on relational databases. Several attempts of peer-to-peer (P2P) architectures and protocols for MMOGs are reviewed, and critical issues such as cheat prevention on P2P MMOGs are highlighted.

Section 4. Security, focuses on the implementation of grid and cloud technologies as these technologies even exacerbate the erosion of trust boundaries in business organizations. Organizations currently struggle to close the security loopholes in the Internet technology and are truly averse to public cloud technology that essentially placates the release of the prized corporate data to the management of third party data centers. Reassuring the corporate world of the security of their data is the central objective of cloud security activities such as the management of domains, services and interoperability. Researchers Chaurasia, Verma and Tomar presented the mechanism for sustaining privacy of a vehicle in a vehicular ad hoc network (VANET) through pseudonym update that could be exploited in the grid and cloud environments. For instance, a VANET allows vehicles on the road to disseminate information as they move. An association can be formed between the physical location of the source vehicle and the transmitted messages. This relationship between the physical vehicle and its identity can breach its privacy. This work exploits the strategy of optimal pseudonym update for maximizing privacy as observed by adversaries with different capabilities. Results of this high performance computing indicate that updating pseudonyms in accordance to the strategy maximizes the privacy of a vehicle in the given situation.

Section 5. Applications, focuses on the growth of software applications in grid and cloud environments, such as software-as-a-service (SaaS), platform-as-a-service (PaaS), infrastructure-as-a-service (IaaS) and others. The architecture of cloud applications is fundamentally different from the conventional application models and is considered to be disruptive in nature (a technology paradigm shift). This fundamental shift in implementation approaches will eliminate software installation, maintenance, deployment, management, and support in the current IT departments of companies. This section features several articles demonstrating the current applications of grid and cloud technologies. Researchers Sahota and Hadjinicolaou, in the article titled “Modeling Scalable Grid Information Services with Colored Petri Nets”, showed how information services can be used to facilitate the discovery of resources and services available to meet user requirements and help tune the performance of the grid. This article models PIndex, which is a grouped peer-to-peer network with Colored Petri Nets (CPNs) for scalable grid information services. Based on the CPN model, a simulator was implemented for PIndex simulation and performance evaluation. In another paper, “Deep Analysis of Enhanced Authentication for Next Generation Networks (NGN)”, researcher Mamdouh Gouda improved the user authentication process in NGN - a telecommunication core. The paper presented an enhanced one-pass AKA procedure that eliminates the repeated steps without affecting the security level in NGN, in addition to reducing the Denial of Service (DoS) attacks. The presented mechanism has minimal impact on the network infrastructure and functionality and does not require any changes to the existing authentication protocols.

Further advances were recorded by researchers Chakrabarty, Collier and Mukhopadhyay in their paper titled “Adaptive Routing Strategy for Large Scale Rearrangeable Symmetric Networks”. The article proposed an adaptive unicast routing algorithm for large scale symmetric networks comprising 2×2 switch elements such as Benes networks. The authors’ algorithm makes routing decisions dependent on the status of each switching element at every stage of the network, hence the name adaptive routing. The approach provides a low complexity solution, but with much better blocking performance than random routing algorithms as demonstrated in the simulation results for various input loads. In another paper titled “G2G: A Meta-Grid Framework for the Convergence of P2P and Grids”, researchers Chung, Hsu, Lin, Lai and Chung proposed a meta-grid framework, named the Grid-to-Grid (G2G) framework, to harmonize autonomic grids in realizing a grid federation. The G2G framework is a decentralized management framework that is built on top of existing autonomic grid systems. This paper further adopts a super-peer network in a separate layer to coordinate distributed grid systems. A super-peer overlay network is constructed for communication among super-peers, thus enabling collaboration among grid systems. The experimental results show that the proposed meta-grid framework can improve system performance with little overhead.

This book presents further applications that feature various uses of grid and cloud technologies. In the paper titled “One Anchor Distance and Angle Based Multi - Hop Adaptive Iterative Localization Algorithm for Wireless Sensor Networks”, researchers Kotwal, Tomar and Abrol presented the distance and angle measurements based on the Multi-Hop Adaptive and Iterative Localization algorithm for localization of unknown nodes in wireless sensor networks (WSNs). The results clearly indicate an improved efficiency of the proposed algorithm in comparison with existing algorithms. In another work -“Fuzzy Allocation of Fine-Grained Compute Resources for Grid Data Streaming Applications”, researchers Zhang, Zhong, Liu and Wu proposed a fuzzy logic controller designed with several simple but robust rules to improve processing efficiency and resource utilization of data streaming applications. Performance of this controller is verified to out-perform classic controllers in response rapidness and less oscillation. The researchers obtained an empirical formula for tuning essential parameters to achieve better performance. Another paper titled “Managing Inconsistencies in Data Grid Environments: A Practical Approach” by Ahmed, Bessis, Norrington and Yue dwelt on one of the main concerns when integrating data from heterogeneous data sources, that is, data redundancy. The researchers take the view that there is a greater chance of data inconsistencies, such as data redundancies when integrating them within a grid environment as compared to traditional distributed paradigms. To this end, the researchers proposed matching strategy between structure objects and data values across federated databases in a grid environment.

Currently, high performance computing (HPC) systems are being integrated into grid and cloud systems as mainframes and supercomputers make a comeback in research and data centers. Computing systems approaching the teraflops-region are considered to be HPC-computers. The networking requirements of grid and cloud systems necessitate the use of a collapsed network backbone for HPC technologies, as the collapsed network backbone architecture is more amenable to easy troubleshooting and single router application. Furthermore, since grids and clouds use multiple processors and computers, scaling problems may cripple critical systems that are retrofitted with HPC technologies, hence the need for further research. In this book series, several chapters feature the application of HPC technologies in scientific research or computational science. Frank and Mesentean dwelt on “Efficient Communication Interfaces for Distributed Energy Resources”, while Mohana and Ashwathakumar harped on “Road Traffic Parameters Estimation by Dynamic Scene Analysis: A Systematic Review”. Further HPC papers are titled as follows: “Intelligent Industrial Data Acquisition and Energy Monitoring using Wireless Sensor Networks” by Gupta, Verma, Tomar and Abrol; “A Method of 3-D Microstructure Reconstruction in the Simulation Model of Cement Hydration” by Dongliang Zhang and “One Anchor Distance and Angle Based Multi - Hop Adaptive Iterative Localization Algorithm for Wireless Sensor Networks” by Kotwal, Tomar and Abrol.

In conclusion, there is a surge in grid and cloud research, as the industry and business world now recognize the potentials in this disruptive technology. The book highlights the direction of current research with sufficient materials to spur efforts in this field by government, industry, institutions and individuals. In that light, this book impacts the advance of cloud technology and global economy, and indeed the information age.

Emmanuel Udoh
National College, Fort Wayne, USA

Author(s)/Editor(s) Biography

Emmanuel Udoh is currently Dean and Professor, College of Information and Computer Technology, Sullivan University, USA. Prior to his current position, Dr. Udoh was the Chair/Director of the IT Department at National College and an Assistant Professor of Computer Science at Indiana University-Purdue University in Fort Wayne. Dr. Udoh holds two doctoral degrees, one in Information Technology from Capella University and one in Geology from Erlangen University in Germany. He also holds an MBA from Capella, an MS in Computer Science from Troy University in Alabama, an MS in Geology from Muenster University in Germany and a BS in Geology from the University of Ife (OAU) in Nigeria. Dr. Udoh is the author of six books and numerous peer-reviewed articles in IT. Dr. Udoh has been listed in American Marquis Who’s Who in the World (1993-1994).