Dynamic multi-agent systems require some accountability mechanism to ensure goal distribution and completion. Accountability can be maintained through proper delineation of power. An organisation is well delineated wh...
详细信息
Dynamic multi-agent systems require some accountability mechanism to ensure goal distribution and completion. Accountability can be maintained through proper delineation of power. An organisation is well delineated when each member has an unambiguous power relationship with its peers. Since responsibility and authority follow fully de-conflicted patterns, a delineated organisation is well suited for solving problems. This paper describes a non-deterministic agent level algorithm that emergently simulates various organisational structures and maintains a well-delineated power structure. The appropriateness of this algorithm is demonstrated in a highly dynamic, poor communication network system. A metric is provided which quantifies the appropriateness of the emerging organisational structures. A structure is considered appropriate if an unambiguous chain of accountability exists leading to the human overseer of the system.
A novel 4D architecture has been presented in [1][3] for the future Internet, which advocates a decomposition of network control functions into three distinct planes: decision, dissemination, and discovery, together w...
详细信息
A novel 4D architecture has been presented in [1][3] for the future Internet, which advocates a decomposition of network control functions into three distinct planes: decision, dissemination, and discovery, together with a streamlined data plane, to achieve more predictable network performance. An underlying concern with the 4D architecture is that the decision making is centralized at a single entity called the Decision Element (DE), whose failure may disrupt the whole network [2]. A simple approach has been proposed to increase the reliability of the system by attaching a few hot stand-by DEs to the network. These hot stand-by DEs collect information from other network elements and perform computation in the same manner as the master DE. In this work, we create a set of experiments to verify the performance of the election algorithm based on the prototype of the 4D architecture and experimentally evaluate the re-convergence time required by the election algorithm after a failure of the master DE.
With the dynamic and heterogeneous characteristics of resources in grid environment, efficient service discovery becomes a challenging issue. In this paper, we propose a service discovery approach based on the distrib...
详细信息
ISBN:
(纸本)1424400600
With the dynamic and heterogeneous characteristics of resources in grid environment, efficient service discovery becomes a challenging issue. In this paper, we propose a service discovery approach based on the distributed spanning-tree architecture. According to this architecture, we arrange the grid resources in such a hierarchical way as IS layer, Institution layer, Organization layer and Domain layer in turn. Any member at the same layer is equivalent. Only IS layer and Institution layer have real resource entities while other layers over them are overlays to cluster the resource information onto their own Delegation Nodes (DN). These DNs have the Resource Index and can locate the query route. Hence, query message traffic is sharply decreased depending on the DN's locating of query routes. Every Institution in grid owns a global unique ID according to the coding mechanism. Based on this, the parallel searching process can be synchronously implemented among peer nodes, which improves the efficiency dramatically. Caching and Collaboration Units techniques are adopted to increase the efficiency of service discovery. The DN can be replaced by the new elected one when it fails which ensures the survivability of the system. Performance evaluation shows that our approach achieves a good efficiency, scalability, survivability and adaptability.
With the dynamic and heterogeneous characteristics of resources in grid environment, efficient service discovery becomes a challenging issue. In this paper, we propose a service discovery approach based on the distrib...
详细信息
With the dynamic and heterogeneous characteristics of resources in grid environment, efficient service discovery becomes a challenging issue. In this paper, we propose a service discovery approach based on the distributed spanning-tree architecture. According to this architecture, we arrange the grid resources in such a hierarchical way as IS layer, Institution layer, Organization layer and Domain layer in turn. Any member at the same layer is equivalent. Only IS layer and Institution layer have real resource entities while other layers over them are overlays to cluster the resource information onto their own Delegation Nodes (DN). These DNs have the Resource Index and can locate the query ***, query message traffic is sharply decreased depending on the DN's locating of query routes. Every Institution in grid owns a global unique ID according to the coding *** on this, the parallel searching process can be synchronously implemented among peer nodes, which improves the efficiency dramatically. Caching and Collaboration Units techniques are adopted to increase the efficienry of service discovery. The DN can be replaced by the new elected one when it fails which ensures the survivability of the system. Performance evaluation shows that our approach achieves a good efficiency, scalability, survivability and adaptability.
Natural distributed systems are adaptive, scalable and fault-tolerant. Emergence science describes how higher-level self-regulatory behaviour arises in natural systems from many participants following simple rule-sets...
详细信息
ISBN:
(纸本)0780377338
Natural distributed systems are adaptive, scalable and fault-tolerant. Emergence science describes how higher-level self-regulatory behaviour arises in natural systems from many participants following simple rule-sets. Emergence advocates simple communication models, autonomy and independence, enhancing robustness and self-stabilization. High-quality distributed applications such as autonomic systems must satisfy the appropriate nonfunctional requirements which include scalability, efficiency, robustness, low-latency and stability. However the traditional design of distributed applications, especially in terms of the communication strategies employed, can introduce compromises between these characteristics. This paper discusses ways in which emergence science can be applied to distributed computing, avoiding some of the compromises associated with traditionally-designed applications. To demonstrate the effectiveness of this paradigm, an emergent election algorithm is described and its performance evaluated The design incorporates non-deterministic behaviour. The resulting algorithm has very low communication complexity, and is simultaneously very stable, scalable and robust.
We improve on Cidon et al.'s leader election algorithm by presenting an algorithm that uses only O(square-root n log D + f) ti-e units to run on synchronous networks of degree f and diameter D, where f greater-tha...
详细信息
We improve on Cidon et al.'s leader election algorithm by presenting an algorithm that uses only O(square-root n log D + f) ti-e units to run on synchronous networks of degree f and diameter D, where f greater-than-or-equal-to 3. When f is 2, the algorithm uses only O(log D) time units.
暂无评论