ArticlePDF Available

Critical Analysis of Dynamic Resource Management for Distributed Systems

March 2011

Authors:

Rajiv Vasantrao Dharaskar

Director, IIIT Kottayam, Kerala, India Institute of National Importance

Swati Sunil Sherekar

Sant Gadge Baba Amravati University

Dynamic Resource Management and deployment of next generation networks (i.e. 4G) begins to spread throughout the world. Open and dynamic environment require flexibility and scalability that can be customized, adopted and reconfigured dynamically, which face the changing environment and requirement. In this paper, the focus is on the current trends, which re used to satisfy the requirement of the Dynamic Resource Management for distributed and wireless systems. The requirements such as distributed OSGi environment, scheduling of multi cell capacity maximization, adaptive middleware for complex heterogeneous distributed systems, self adaptive context processing framework for wireless sensor network, dynamic resource allocation for targeted throughput and flexible, adaptive security middleware.

Content uploaded by Rajiv Vasantrao Dharaskar

Content may be subject to copyright.

ResearchGate has not been able to resolve any citations for this publication.

ResearchGate has not been able to resolve any references for this publication.

Conference Paper

When Collective Dynamic Knowledge is Greater Than the Sum of Its Parts: the Case for Mobile Wireless...

January 2008

The use of wireless sensor networks is one of the most rapidly growing areas because of their broad applications - from medical care to environmental monitoring to military surveillance. This review paper surveys the use of collaborative mobile robotic agents in wireless sensor networks to improve detection and tracking performance. We first identify five major conditions influencing the ... [Show full abstract] applications in wireless sensor networks: situation awareness, priorities and metrics for goals or tasks, control and optimization, location and distribution of computing power, and location and distribution of electrical power. We then present a summary of design topics: coverage and deployment, localization, data fusion, communication protocols, and energy consumption and management. Anticipated research directions are also discussed in the conclusions.

Conference Paper

Using the TrustME Tool Suite for Automatic Component Protocol Adaptation

April 2002 · Lecture Notes in Computer Science

The deployment of component oriented software approaches gains increasing importance in the computational sciences. Not only the promised increase of reuse makes components attractive, but also the possibilities of integrating different stand-alone programs into a distributed application. Middleware platforms facilitate the development of distributed applications by providing services and ... [Show full abstract] infrastructure. Component developers can thus benefit from a common standard to shape components towards and application designers from using pre-fabricated software components and shared platform services. Although such platforms claim to achieve fast and flexible development of distributed systems, they fall short in key requirements to reliability and interoperability in loosely coupled distributed systems. For example, many interoperability errors remain undetected during development and the adaptation and integration of of third-party components still requires major effort and cost. Partly this problem can be alleviated by the use of formal approaches to automatic interoperability checks and component adaptation. Our Reliable Architecture Description Language (RADL) is aimed at precisely this problem. In this paper we present key aspects of RADL used to specify component-based, compositional views of distributed applications. RADL involves a rich component model, enabling protocol information to be contained in interfaces. We focus on protocol-based notions of interoperability and adaptation, important for the construction of distributed systems with loosely coupled components.

Article

A new development environment for an event-based distributed system

August 2003 · Computer Standards & Interfaces

The rapid growth of data exchange on the Internet has created many critical problems that require an answer. Traditional data exchange systems based on client/server communication models are less scalable and incur especially high maintenance cost in the data exchange domain. For these reasons, many researchers have switched their interest to asynchronous communication models. Although ... [Show full abstract] Message-Oriented Middleware (MOM) is a middle-tier infrastructure that links operating systems and applications, such asynchronous communication APIs supported by middleware vendors are usually hard to use. For these reasons, in this study, we present a new development environment for asynchronous communication platforms which we term Ghostwriter. The keyword for our development environment is ‘easy’, that is, easy to use, easy to develop, and easy to deploy. Therefore, many researchers have switched their interest to asynchronous communication models. In addition, learning about and implementing the functions of the asynchronous communication's clients in Ghostwriter environment is simple. Other benefits are a lower technical learning curve, help for concentrate on system design, has easily reusable components, and easily integrated applications.

Thesis

Full-text available

Dynamic resource allocation and management in virtual networks and Cloud.

December 2015

Houda Jmila

Cloud computing is a promising technology enabling IT resources reservation and utilization on a pay-as-you-go manner. In addition to the traditional computing resources, cloud tenants expect compete networking of their dedicated resources to easily deploy network functions and services. They need to manage an entire Virtual Network (VN) or infrastructure. Thus, Cloud providers should deploy ... [Show full abstract] dynamic and adaptive resource provisioning solutions to allocate virtual networks that reflect the time-varying needs of Cloud-hosted applications. Prior work on virtual network resource provisioning only focused on the problem of mapping the virtual nodes and links composing a virtual network request to the substrate network nodes and paths, known as the Virtual network embedding (VNE) problem. Little attention was paid to the resource management of the allocated resources to continuously meet the varying demands of embedded virtual networks and to ensure efficient substrate resource utilization. The aim of this thesis is to enable dynamic and preventive virtual network resources provisioning to deal with demand fluctuation during the virtual network lifetime, and to enhance the substrate resources usage. To reach these goals, the thesis proposes adaptive resource allocation algorithms for evolving virtual network requests. First, we will study in depth the extension of a virtual node, i.e. an embedded virtual node requiring more resources, when the hosting substrate node does not have enough available resources. Second, we will improve the previous proposal to consider the substrate network profitability. And finally we will deal with the bandwidth demand variation in embedded virtual links. Consequently, the first part of this thesis provides a heuristic algorithm that deals with virtual nodes demand fluctuations. The main idea of the algorithm is to re-allocate one or more co-located virtual nodes from the substrate node, hosting the evolving node, to free resources (or make room) for the evolving node. In addition to minimizing the re-allocation cost, our proposal takes into account and reduces the service interruption during migration. The previous algorithm was extended to design a preventive re-configuration scheme to enhance substrate network profitability. In fact, our proposal takes advantage of the resource demand perturbation to tidy up the SN at minimum cost and disruptions. When re-allocating virtual nodes to make room for the extending node, we shift the most congested virtual links to less saturated substrate resources to balance the load among the SN. Our proposal offers the best tradeoff between re-allocation cost and load balancing performance. Finally, a distributed, local-view and parallel framework was devised to handle all forms of bandwidth demand fluctuations of the embedded virtual links. It is composed of a Controller and three algorithms running in each substrate node in a distributed and parallel manner. The framework is based on the self-stabilization approach, and can manage many and different forms of bandwidth demand variations simultaneously.