In this paper, the classification of nonoscillatory solutions for a two-dimensional neutral difference system is considered. Sufficient and/or necessary conditions of existence for those solutions will be established.
ISBN:
(纸本)0769529097
In this paper, the classification of nonoscillatory solutions for a two-dimensional neutral difference system is considered. Sufficient and/or necessary conditions of existence for those solutions will be established.
This paper proposes a novel algorithm for load balancing in the parallel and distributed systems. Firstly, almost all jobs are organized into the standard job combinations, each of which consists of one to four jobs. ...
详细信息
ISBN:
(纸本)0769529097
This paper proposes a novel algorithm for load balancing in the parallel and distributed systems. Firstly, almost all jobs are organized into the standard job combinations, each of which consists of one to four jobs. Then they are assigned to the machines according to the assignment algorithm for job combinations, which is a special integer partition algorithm. The maximum error between the optimal algorithm and our algorithm is estimated. The experimental results show that our algorithm performs efficiently.
Minimizing communication by increasing the locality of data references is an important optimization for achieving high performance on distributed memory machines. But in the progress of decomposition, reorganization i...
详细信息
ISBN:
(纸本)9780769529097
Minimizing communication by increasing the locality of data references is an important optimization for achieving high performance on distributed memory machines. But in the progress of decomposition, reorganization is inevitable. And the communication produced by reorganization is inevitable too. In this paper, the authors present a linear decomposition algorithm that automatically finding computation and data decomposition, including finding data and computations decomposition that has data reorganization communication. And the authors improve the method and reduce the communication cost by merging parallel regions with the same data decomposition.
A large number of resource locating algorithms have been proposed in peer-to-peer (P2P) networks. However, the query efficiency of most existing methods is limited to large-scale P2P networks. In this paper, we propos...
详细信息
ISBN:
(纸本)0769529097
A large number of resource locating algorithms have been proposed in peer-to-peer (P2P) networks. However, the query efficiency of most existing methods is limited to large-scale P2P networks. In this paper, we propose a new decentralized P2P architecture, by taking advantage of the heterogeneous characters of nodes to integrate the resources of stable nodes, and simplify the management of dynamic nodes based on two different types of clustering. In our proposed architecture, group-based semantic overlay network is formed by node clustering and resources of stable nodes are organized by a set of cooperative resource clustering frameworks. When a query is generated, it will be first processed in local group, and if it fails to hit the targets, it will be relayed to the resource clustering frameworks. simulation shows that this architecture can greatly reduce the search traffic in large-scale P2P networks.
Program verification is a major concern and challenge to our highly computer-dependent societies. Under a closed environment, this paper explores proof reusing and organizing Methodology. After making some extension t...
详细信息
ISBN:
(纸本)0769529097
Program verification is a major concern and challenge to our highly computer-dependent societies. Under a closed environment, this paper explores proof reusing and organizing Methodology. After making some extension to Hoare's convention, and defining the verification task we present a model suitable for verifying numerous programs simultaneously.
A problem of research on overlay network is lack of testing platform. Recent research has shown that one can use distributed Hash Tables (DHTs) to build scalable, robust and efficient applications. However, large-scal...
详细信息
ISBN:
(纸本)9780769529097
A problem of research on overlay network is lack of testing platform. Recent research has shown that one can use distributed Hash Tables (DHTs) to build scalable, robust and efficient applications. However, large-scale distributed systems are hard to deploy, nor are DHTs. OpenDHT is a publicly accessible DHT service. In contrast to the usual DHT model, clients of OpenDHT can issue put and get operations to any DHT node, which processes the operations on their behalf. This paper introduces OpenDHT and concentrates on how to do the network testing based on OpenDHT. The testing about resource location is also expounded.
To test the validity of the eigenfunction expansion method (EEM), a great deal of wave profiles calculated by EEM are compared with the corresponding results obtained by the boundary element method (BEM), which has be...
详细信息
ISBN:
(纸本)0769529097
To test the validity of the eigenfunction expansion method (EEM), a great deal of wave profiles calculated by EEM are compared with the corresponding results obtained by the boundary element method (BEM), which has been verified by many experiments. The comparisons shown that the relative errors of the maximum wave height between them are smaller than 5.5% after 10 periods when the initial wave steepness is below 0.157. However, the errors exceed 12.3% if the initial wave steepness greater than 0.198. As references, we also compared them with the linear solutions. The investigation confirmed that EEM is only valid in the transient wave propagation simulation for small or moderate wave steepness as expected. The corresponding spectrum analysis also reveals that EEM fails to capture harmonics with high wave numbers for the steep waves.
The Local Time Warp (LTW) model was one of the novel approaches to distributedsimulation presented in early 90s. While several similar schemes had emerged afterwards, no implementation and performance analysis of LTW...
详细信息
ISBN:
(纸本)9780769528984
The Local Time Warp (LTW) model was one of the novel approaches to distributedsimulation presented in early 90s. While several similar schemes had emerged afterwards, no implementation and performance analysis of LTW were offered. This paper provides an implementation and performance analysis of the Local Time Warp model on a Beowulf Cluster using Message Passing Interface (MPI). Cluster computing matches well with the Local Time Warp model due to its partitioning of the simulation entities and assigning them to clusters of processors. This paper presents a distributedsimulation engine called BG-LTW which is a clustered-based implementation of the model. Further, the implementation features and its performance are provided. The obtained results suggest that for large-scale simulations local time warp outperforms its global counterpart.
simulation is a low cost and safe alternative to solve complex problems in various areas. To promote reuse and interoperability of simulation applications and link geographically dispersed simulation components, distr...
详细信息
ISBN:
(纸本)9780769528984
simulation is a low cost and safe alternative to solve complex problems in various areas. To promote reuse and interoperability of simulation applications and link geographically dispersed simulation components, distributedsimulation was introduced. The High Level Architecture (HLA) is the IEEE standard for distributedsimulation. To optimize communication efficiency between simulation components, HLA defines a Data Distribution Management (DDM) service group for filtering out unnecessary data exchange. It relies on the computation of overlap between update and subscription regions, which is called matching. In this paper we propose an efficient sort-based DDM matching algorithm for HLA applications with a large spatial environment. A theoretical analysis of our algorithm concludes that it should have good storage and computational scalability. The experimental results have verified the theoretical conclusions by showing that our algorithm has much less storage requirement than the original sort-based matching algorithm and generally has the best computational performance when compared with region-based and the original sort-based matching algorithms.
Metadata management is a key technique of data grid. This paper presents a replication and cache based metadata management system (RCMMS) to improve metadata management of Global distributed Storage System (GDSS). Sto...
详细信息
ISBN:
(纸本)0769529097
Metadata management is a key technique of data grid. This paper presents a replication and cache based metadata management system (RCMMS) to improve metadata management of Global distributed Storage System (GDSS). Storage Set-vice Provider (SSP) of GDSS has dedicated buffer to cache metadata. The design, implementation and evaluation of RCMMS are discussed in this paper. RCMMS provides efficient algorithms to manage highly dynamic metadata replicas. The evaluation demonstrates that replica plus cache metadata management system outperforms existing metadata management system of GDSS.
暂无评论