In this paper, we show that computing the distributed program reliability on the star distributedcomputing systems is NP-hard. We develop a polynomially solvable case to compute distributed program reliability when s...
详细信息
In this paper, we show that computing the distributed program reliability on the star distributedcomputing systems is NP-hard. We develop a polynomially solvable case to compute distributed program reliability when some additional file distribution is restricted on the star topology. We also propose a polynomial time algorithm for computingdistributed program reliability with approximate solution when the star topology is not satisfied with the additional file distribution.
The main obstacles to grid file management come from the fact that grid file resources are typically stored in heterogeneous and distributed environment and accessed through various protocols. In this paper, we propos...
详细信息
ISBN:
(纸本)3540366792
The main obstacles to grid file management come from the fact that grid file resources are typically stored in heterogeneous and distributed environment and accessed through various protocols. In this paper, we propose a grid file management system called Vega [1][2] Hotfile2 for data-intensive application in widely distributed systems and grid environments. Widely distributed and heterogeneous storage resource can converge into a single uniform storage resource using virtualization techniques in Hotfile2. It is a grid middleware provides a uniform file resources view and a high level interface to access widely distributed and heterogeneous data resources. Presented here are two services that we believe are fundamental to any grid file management system: high-performance, secure transport and metadata management. In this paper, we discuss many aspects of our design and implementation in Hotfile2, present the file system interface designed to support distributed applications such as batch job. Experiments are done to illustrate its performance, extensibility, and various other features.
Counting networks are well studied parallel and distributed data structures, which are useful in synchronization applications such as distributed counting and load balancing. However current constructions of counting ...
详细信息
ISBN:
(纸本)0769523315
Counting networks are well studied parallel and distributed data structures, which are useful in synchronization applications such as distributed counting and load balancing. However current constructions of counting networks are static, since their width (the degree of parallelism), and hence the size of the network, have to be fixed in advance. This present an obstacle in efficiently implementing them in a large distributed system whose size may be changing, due to nodes joining and leaving the network. We present an adaptive construction of the bitonic counting network. Our network tunes its width to the system size in a distributed and local way. With high probability, the effective "width" of the network is Omega(N/log(2) N), where N is the number of nodes currently in the system, and the effective "depth" of the network is O(log(2) N). In contrast, a static implementation would have the same width irrespective of the system size. When the system size changes, the network adapts by splitting or merging its components. All decisions and actions are decentralized: these include the decision of when to split and merge the components, and the action of splitting and merging them. Our construction is layered on an overlay network which provides an efficient peer-to-peer lookup service, and uses the recursive structure present in the bitonic network to adapt its implementation. Though we discuss the bitonic network, our technique could be applied to build an adaptive implementation of any distributed data structure which can be decomposed in a recursive way.
A computational grid is a widespread computing environment that provides huge computational power for large-scale distributed applications. Load balancing, has a considerable effect on the grid middleware performance....
详细信息
ISBN:
(纸本)9783540681397
A computational grid is a widespread computing environment that provides huge computational power for large-scale distributed applications. Load balancing, has a considerable effect on the grid middleware performance. Current load balancing methods cannot satisfy all necessities for the grid. In this paper, a Multi-level Load Balancing Method (MLBM) is proposed. Cooperation among different levels in this method, removes disadvantages of each level, while satisfy most of load balancing requirements needed. Simulation results indicate that this new mechanism surpasses its predecessors in increasing efficiency and decreasing communication overhead.
Computational grids consist of distributed environments where partner institutions offer hosts along with computational resources that can be used by all members of the grid. When an application needs to run in such e...
详细信息
ISBN:
(纸本)9789897583650
Computational grids consist of distributed environments where partner institutions offer hosts along with computational resources that can be used by all members of the grid. When an application needs to run in such environment, it allocates a portion of hosts necessary for its executions. Traditionally, the workload imposed on computational grids has a characteristic of being bag-of-tasks (BoT). It means that multiple replicas are submitted to different hosts, and when a response is processed, such replicas are either ignored or released. On resource allocation, as the grid is distributed among different participants, only idle resources can be leased by a new application. However, due to the behavior of BoTs, many allocated resources do not use their resources in their entirety. Another important fact is that only fully idle hosts can be added to the grid pool, and used only at these times. From the above, this paper proposes an approach that uses underutilized resource slice of grid hosts through virtualization, adjusting the use of grid applications to the leftover resources from daily hosts usage. It allows grid applications to run, even when hosts are in use, as long as there is an idle slice of resources, and their use does not interfere with the host's current running. To evaluate this approach, we performed an empirical evaluation of a virtualized server running applications concurrently with a virtualized grid application. The results have showed that our scheme could accelerate the performance of grid applications without impacting on higher energy consumption.
Partitioning meshes is a preprocessing step for parallel scientific simulation. The quality of a partitioning is measured by load balance and communication overhead. The effectiveness of a partitioning significantly i...
详细信息
ISBN:
(纸本)0769516777
Partitioning meshes is a preprocessing step for parallel scientific simulation. The quality of a partitioning is measured by load balance and communication overhead. The effectiveness of a partitioning significantly influences the performance of parallel computation. In this paper, we propose a quadtree spatial-based domain decomposition method for partitioning unstructured meshes. The background quadtree, which is originally used to represent the density distribution among elements within the computing domain, can be used to obtain an initial partitioning and to do multi-level refinement. As the quadtree implicitly defines hierarchical relationship, which is a natural way to define coarsening and uncoarsening phases, we can repeatedly apply coarsening, partitioning, and uncoarsening multi-level refinement phases, until no improvement can be made. Thus, for most cases, the partitioning results by our method are better than those produced by other graph-based partitioning methods. Experimental studies for the NACA0012 airfoil, the NASA EET wing, and an artillery shell within a shock tube are reported.
Diagnosing exceptions in multi-agent systems (MAS) is a complex task due to the distributed nature of the data and control in such systems. This complexity is exacerbated in open environments where independently devel...
详细信息
ISBN:
(纸本)0780385667
Diagnosing exceptions in multi-agent systems (MAS) is a complex task due to the distributed nature of the data and control in such systems. This complexity is exacerbated in open environments where independently developed autonomous agents interact with each other in order to achieve their goals. Inevitably, exceptions will occur in such MAS and these exceptions can arise at one of three levels, namely environmental, knowledge and social levels. In this paper we propose a novel exception diagnosis system that is able to analyse and detect exceptions effectively. The proposed architecture consists of specialised exception diagnosis agents called sentinel agents. The sentinel agents are equipped with knowledge of observable abnormal situations, their underlying causes, and resolution strategies associated with these causes. The sentinel agent applies a heuristic classification approach to collect related data from affected agents in order to uncover the underlying causes of the observed symptoms. We illustrate and evaluate our proposed architecture using an agent-based gridcomputing case study.
Most Lipschitzian Global Optimization algorithms perform an exhaustive search using a branch-and-bound (B&B) scheme. The question is how to run multi-dimensional Lipschitz Global Optimization in parallel, such tha...
详细信息
ISBN:
(纸本)9780769550947
Most Lipschitzian Global Optimization algorithms perform an exhaustive search using a branch-and-bound (B&B) scheme. The question is how to run multi-dimensional Lipschitz Global Optimization in parallel, such that the implementation depending on the used platform is efficient. Previous work shows a parallel version developed for multicore nodes with two levels of parallelism: intra-node and inter-node. On intra-node level, one can perform dynamic load balancing by generating threads dynamically. Threads end when they complete their assigned work. The inter-node level carries out a static load balancing using MPI. In general, algorithm design depends on the characteristics of problems to be solved. There are several ways to improve performance of general parallel B&B algorithms. Specifically, we are interested in how to apply them to parallel Lipschitz Global Optimization algorithms. Operations like selecting the next subproblem to be evaluated become critical in parallel B&B schemes. We study Depth and Hybrid (Best-Depth) options as selection criterion. Previous work, using only MPI or OpenMP, discard not only the broadcasting of the best found upper bound of the solution due to its high average cost/performance but also the use of dynamic load balancing. Here we check how broadcasting the upper bound affects the developed MPI-Pthreads algorithm. Additionally, we study how to perform dynamic load balancing at inter-node level. Experimental results show which designs perform better on which type of instances for the used computational architecture.
Power can be generated from conventional as well as non-conventional sources. Power can be accepted from any type of source. With the increasing share of generating resources adding up in distribution side, the tradit...
详细信息
ISBN:
(纸本)9781509058389
Power can be generated from conventional as well as non-conventional sources. Power can be accepted from any type of source. With the increasing share of generating resources adding up in distribution side, the traditional frontier that is separating all the three parts (generation, transmission and distribution) of power system are vanishing rapidly. As the grid is transforming to microgrid and smart grid, the consumers are now termed as "prosumers". The grid that is initially designed for one-way power flow is now experiencing two-way flow of power. As the energy storage cost is declining, it gives rise to new debate for the power system intellectualist whether to invest in conventional grid model (generation + transmission + distribution) or to invest in storage model (distributed generation + distribution + storage) With evolution of Microgrid and Smartgrid concept, power system is advancing to a new era of digitalized power system. With the digitalization of power system there comes various pros and cons in terms of technical as Well as economic feasibility. Thus, requires new researches in this modern field.
暂无评论