Resource and application-aware resource discovery in computing environments

In this paper, we establish a relationship between resource discovery protocol and cluster head election procedure. While doing so, we propose a multi-resource-criteria cluster head election mechanism that reduces the computational and communication overhead incurred in resource discovery. The consequent resource discovery protocol thus necessitates a minimalist exchange of information that we provide through custom packet formats and message flows. The extent of improvement thus achieved in the proposed protocol over existing resource discovery mechanisms is shown by performance evaluation.

Resource discovery in disconnected delay-tolerant mobile social networks (MSNs) continues to be a challenging issue. Nowadays, the resource discovery mechanisms for MSNs are classified into three categories: social connectivity-based, movement pattern-based and content similarity-based resource discovery mechanisms. The first two categories have some advantages, such as low transmission delay and management overhead, the deficiency of the two mechanisms is that the poorer searching efficiency without considering the content similarity. The third category facilitates the higher searching efficiency. Yet, it leads to higher traffic cost due to not considering the physical location information of nodes. Furthermore, the current methods are designed in Cartesian coordinates, which is not suitable for real physical network environments. In this paper, we propose a novel resource discovery mechanism in 3-dimensional Cartesian coordinate system. Our scheme fully utilizes the content similarity, the movement pattern and the location awareness to achieve the higher searching efficiency and reduce the system overhead for MSNs. We simulated our method on the platform of the opportunistic network environment simulator. Simulations results show that our proposed scheme outperforms the state-of-the-art resource discovery schemes in terms of the searching efficiency and the average delay.

In ubiquitous stub network, it is a critical challenge to realize resource discovery and share under disconnected network topology. In this paper, a cluster-based resource discovery mechanism is proposed with resource registration, distribution and routing model. Firstly, we use resource directory index nodes to assist in resource management. Secondly, we use inter-cluster mobile terminals to support resource routing. In addition, we take the nodes contact probability into account and establish the minimum expectation delay routing standard to opportunistically route between terminals. At last, the simulation result shows this mechanism is better applied to support disconnected ubiquitous resource discovery.

In large-scale computational or P2P Grids, discovery of heterogeneous resources as a working group is crucial to achieving scalable performance. This paper presents a hierarchical cycloid overlay (HCO) architecture with resource clustering and discovery algorithms for efficient and robust resource discovery in wide-area distributed Grid systems. We establish program/data locality by clustering resources based on their physical proximity and functional matching with user applications. We further develop randomized probing and cluster-token forwarding algorithms. The novelty of the HCO scheme lies in low overhead, fast speed and dynamism resilience in multi-resource discovery. The paper presents the HCO framework, new performance metrics, and simulation experimental results. This HCO scheme compares favorably with other resource management methods in static and dynamic Grid applications. In particular, it supports efficient resource clustering, reduces communications cost, and enhances resource discovery success rate in promoting large-scale distributed supercomputing applications.

A survey on resource discovery mechanisms, peer-to-peer and service discovery frameworks

Resource discovery mechanisms are key for distributed computing. Traditional approaches use dedicated servers to store information about the whole system and make all the scheduling decisions. Many mechanisms exist for decentralized environments, based on distributing the information in a certain way over a structured overlay network. This usually creates complex relationships between the distribution of the data to be stored and the performance of the system. In this paper, we present a method for resource discovery with the simplicity and power of centralized approaches, designed for decentralized systems formed by non-dedicated resources. We show that it provides a constant look-up latency for frequent resources in the average case, while causing a low overhead and tolerating the levels of churn that are to be expected from non-dedicated resources.

Grid is commonly considered as next generation network, Resource discovery is the key requirements in large heterogeneous grid environments, and an effective and efficient resource discovery mechanism is crucial. Base on centering resource discovery and distributed resource discovery methods, a hierarchical resource discovery method is proposed. Agent technology is used in every layer. Resource adjacency list as a kind of centering resource discovery method is used in the Supernode of every VO .But in the middle of the SuperNodes of P2P layer, we use Ant Colony algrithm as distributed resource discovery method.

Multitude of resources pooled in grid environments require a scalable, efficient, low-cost and self-organizing resource discovery mechanisms. P2P networks possess overlapping characteristics like large scale resource sharing and dynamic nature of participation. P2P style resource discovery approaches have been employed in the grid environments. We propose a P2P based resource discovery approach with modified features like customized neighbourhood defined according to the user preferences and distributed search using directed flow of requests contrary to the blind propagation of P2P systems.

This paper researches on how to realize fast spatial resource discovery in distributed, heterogeneous Grid GIS environment. The characteristic of spatial data as multi-resource, huge volume, heterogeneous formats and time relativity decides that resource discovery in Grid environment is a very complex and exigent problem. Combining the central control with distributed management technology, this paper supposes a two-double resource discovery mechanism with both distributed resource catalog and dynamic hash table, which is just fit to the spatial data access characteristics that distributed in wide area but concentrated in local area. This method overcomes the single node bottleneck by setting up the dynamic service sections among middle layer metadata servers, and utilizes the improved DHT technology to route queries in high efficiency. With the new method, people multi-attribute resource discovery could be supported that not limited by the special keyword as P2P does. It is helpful to solve the spatial information discovery problem in complex environment, which could also enhance the extendibility and credibility of system.

Grid computing refers to the sharing of geographically distributed resources in order to accomplish a desired task. Grids are more secure and enable the job to be broken into chunks, each being executed in a parallel way. Computational Grids serve the purpose of processing a submitted job by finding the required resource from the grid. In implementing computational grids, resource discovery and task scheduling play a major role. Discovery of resources should be done at a minimum cost, avoiding the traversal to unnecessary nodes. Many approaches have been proposed in this area. In this paper, we propose the use of a three-layered, hierarchical, weighted tree structure with bitmaps representing the resources available at each node updated in a way to help the user to make a multi-attribute query. The higher level nodes are put together to form a structured Hyper Cube, resulting in a peer-to-peer fault tolerant and scalable system. Our proposed algorithm outperforms the other resource discovery mechanisms by drastically reducing the number of hops required and also avoiding traversal to unnecessary nodes in the tree. It is a more dynamic and fault tolerant resource discovery mechanism for peer-to-peer grids.

Resource discovery is one of the most important services that significantly affects the efficiency of grid computing systems. The inherent dynamic and large-scale characteristics of grid environments make their resource discovery a challenging task. In recent years, different approaches have been proposed for resource discovery, attempting to tackle the challenges of grid environments and improve the efficiency. Being aware of these challenges and approaches is worthwhile in order to choose an appropriate approach according to the application in different organizations. This study reviews the most important factors that should be considered and challenges to be tackled in order to develop an efficient grid resource discovery system.

Resource discovery is an important part of distributed computing and resource sharing systems, like grids and utility computing. Because of the increasing importance of decentralized and peer-to-peer environments, characterized by high dynamism and churn, a number of resource discovery mechanisms, mainly based on peer-to-peer techniques, have been presented recently. We present and classify them according to criteria like their topology and the degree of achievement of various common requirements of great importance for the targeted environments, as well as compare their reported performance. These classifications intend to provide an intuitive vision of the strengths and weaknesses of each system.

A new type of resource discovery mechanism is proposed for simulation oriented data grid. By introducing distributed resource brokers at edge routers of each domain, a majority of resource discovery queries can be processed solely by resource brokers at edge routers, instead of accessing remote global resource broker. The response time and network overhead of resource discovery is reduced and system scalability is enhanced. Experiment results show that the proposed resource discovery mechanism is feasible.KeywordsData GridGrid ResourceResource ProviderResource AccessResource DiscoveryThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Grid computing is a hot research direction in computer field. Resource management is an important part of this research. In Grid environment, satisfactory services for discovering resources and allocating appropriate resources can be provided to the Grid users though the agents. Agent based resource management with alternate solution (ABRMAS) has been proposed. But the alternate solution is not explicit. This paper adds the middleware grid architecture for computational economy (GRACE) with resources pricing fluctuation manager (RPFM) into ABRMAS. The new model is designed so as to provide negotiate solution when resource discovery fails of nodes occurrence. Through simulation experiments, the system profit and successful rate are all improved by the new model.

Integration of mobile resources in grid environment increases the complexity of resource discovery due to heterogeneity, dynamic nature of resources, mobility pattern and uncertainty in the availability. The mobile grid has geographically distributed autonomous resources and is composed of hardware and software resources that need to be located. The existing resource discovery approaches in mobile grid computing treat resources equally. These approaches face problems in terms of response times and the need to generate complex queries. This can be reduced by grouping of resources based on its characteristics. The dynamic characteristics of mobile resources increase complexities in resource grouping and filtering process, which in turn increases resource discovery time. In this paper, ontology based resource discovery mechanism is proposed for the mobile grid environment. This mechanism uses ontology to group the resources and a map per to filter the resources based on resource information which reduces resource discovery time.