Assume that each vertex of an arbitrary n-vertex tree T is assigned a nonnegative integer weight. This paper considers partitioning the vertices of tree T into p disjoint clusters such that the total weight of each cl...
详细信息
ISBN:
(纸本)9780769548791
Assume that each vertex of an arbitrary n-vertex tree T is assigned a nonnegative integer weight. This paper considers partitioning the vertices of tree T into p disjoint clusters such that the total weight of each cluster is at least l and at most u, where l and u are given integers with l
The proceedings contain 109 papers. The topics discussed include: supporting cache locality optimization with a toolset;model-based performance diagnosis of master-worker parallel computations;tuning application in a ...
详细信息
ISBN:
(纸本)3540377832
The proceedings contain 109 papers. The topics discussed include: supporting cache locality optimization with a toolset;model-based performance diagnosis of master-worker parallel computations;tuning application in a multi-cluster environment;analysis of the memory registration process in the Mellanox Infiniband software stack;an adaptive scheduling method for gridcomputing, improvement of the efficiency of genetic algorithms for scalable parallel graph partitioning in a multi-level framework;multi-dimensional kernel generation for loop nest software pipelining;supporting a real-time distributed intrusion detection application on GATES;design and effectiveness of small-sized decoupled dispatch queues;run-time switching between total order algorithms;data parallel iterators for hierarchical grid and tree algorithms;and a software framework for the portable parallelization of a particle-mesh simulations.
GPGPU (General Purpose computing on Graphics Processing Units) has marked a revolution in the field of parallelcomputing allowing to achieve computational performance unimaginable until a few years ago. This hardware...
详细信息
ISBN:
(纸本)9781467394734
GPGPU (General Purpose computing on Graphics Processing Units) has marked a revolution in the field of parallelcomputing allowing to achieve computational performance unimaginable until a few years ago. This hardware has proven to be extremely reliable and suitable to simulate Cellular Automata (CA) models for modeling complex systems whose evolution can be described in terms of local interactions. Starting from previous GPGPU implementations of CA models with CUDA, this paper presents an effective implementation of a well-known numerical model for simulating lava flows on Graphical Processing Units (GPU) based on the OpenCL (Open computing Language) standard. In addition, a preliminary Civil Defence application related Hazard maps of an area located at Mt. Etna volcano (South Italy), confirms the validity of OpenCL and both low-cost and high-end graphics hardware as an alternative to expensive solutions for the simulation of CA models.
This paper describes a new prolocol for authentication in Peer-to-Peer systems. The protocol has been designed to meet specialized requirements of P2P systems, such as lack of direct communication between peers or req...
详细信息
ISBN:
(纸本)0769524052
This paper describes a new prolocol for authentication in Peer-to-Peer systems. The protocol has been designed to meet specialized requirements of P2P systems, such as lack of direct communication between peers or requirements for controlled anonymity. At the same time, a P2P authentication protocol must be resistant to spoofing, eavesdropping and playback, and man-in-the-middle attacks. The protocol is studied for a model P2P storage system that needs to implement file access rights.
Data replication is a common method used to improve the performance of data access in distributed systems. In this paper, two dynamic replication algorithms, Simple Bottom-Up (SBU) and Aggregate Bottom-Up (ABU), are p...
详细信息
Data replication is a common method used to improve the performance of data access in distributed systems. In this paper, two dynamic replication algorithms, Simple Bottom-Up (SBU) and Aggregate Bottom-Up (ABU), are proposed for the multitier Data grid. A multi-tier Data grid simulator called DRepSim is developed for studying the performances of the dynamic replication algorithms. The simulation results show that both algorithms can reduce the average response time of data access greatly compared to the static replication method. ABU can achieve great performance improvements for all access patterns even if the available storage size of the replication server is very small. Comparing the two algorithms to Fast Spread dynamic replication strategy, ABU proves to be superior. As for SBU, although the average response time of Fast Spread is better in most cases, Fast Spread's replication frequency is too high to be applicable in the real world. (c) 2004 Elsevier B.V. All rights reserved.
Data is dynamically structured by nature and can be highly diverse and multifaceted. Often, such diverse and complex information needs to be linked. Conventional datastores, such as relational databases, do not conven...
详细信息
ISBN:
(纸本)9780769530642
Data is dynamically structured by nature and can be highly diverse and multifaceted. Often, such diverse and complex information needs to be linked. Conventional datastores, such as relational databases, do not conveniently accommodate dynamically varying structures, as frequently modifying database schemas is not feasible. RDF triplestores offer a flexible solution for handling such data, where any property about an entity can be described by a triple having a subject, a predicate, and an object. Also, data is inherently distributed due to origination points, ownership and many other reasons. Furthermore, storing data in triplestores gives rise to the need to distribute data due to the large number of triples that would result by migrating existing data from a database, for example. In this paper we present our work on designing index structures in order to facilitate efficient querying of a distributed triplestore (DTS). The distributed querying algorithm in DTS makes use of a sub-graph isomorphism approach to eliminate traversing edges between triplestores that does not have the potential to produce any results. We show that our triplestore has equivalent performance as 3Store when used in a non-distributed mode. Our performance tests in the distributed mode show that the indexes improve efficiency of querying.
作者:
Tang, LLLi, QQWuhan Univ
LIESMRS State Key Lab Informat Engn Surveying Wuhan Peoples R China Wuhan Univ
Res Ctr Spatial Informat & Network Commun Wuhan Peoples R China
It is publicly considered that the next generational Internet technology is gridcomputing, which supports the sharing and coordinated use of diverse resources in dynamic virtual organizations from geographically and ...
详细信息
ISBN:
(纸本)3540308563
It is publicly considered that the next generational Internet technology is gridcomputing, which supports the sharing and coordinated use of diverse resources in dynamic virtual organizations from geographically and organizationally distributed components. gridcomputing characters strong computing ability and broad width information exchange[1]. Globus presented Open grid Services Architecture (OGSA), which centered on grid services[3]. According to the characteristic of grid-based Virtual Reality (GVR) and the development of current gridcomputing, this paper put forward the Orient-griddistributed Network Model for GVR, whose dynamic Virtual group is corresponding with the Virtual Organization in OGSA service, The GDNM is of more advantage to the distributed database consistency management, and is more convenient to the virtual group users acquiring the GVR data information, and the dynamic virtual groups in GDNM are easier and more directly to utilize the grid source and communication each other. The architecture of GVR designed in this paper is based on OGSA and web services, which is based on the OGSA. This architecture is more convenient to utilizing grid service and realizing the GVR. This paper put forward the method of virtual environment Object-oriented Dynamic Modeling (OODM) based on Problem Solving (PS), which is applied with dynamic digital terrain and dynamic object modeling. This paper presents the implementation of GVR and the interfaces of grid Service.
Science collaborations such as ATLAS at the high-energy particle accelerator at CERN use a computer grid to run expensive computational tasks on massive, distributed data sets. Dealing with big data on a grid demands ...
详细信息
ISBN:
(纸本)9783031087578;9783031087561
Science collaborations such as ATLAS at the high-energy particle accelerator at CERN use a computer grid to run expensive computational tasks on massive, distributed data sets. Dealing with big data on a grid demands workload management and data allocation to maintain a continuous workflow. Data allocation in a computer grid necessitates some data placement policy that is conditioned on the resources of the system and the usage of data. In part, automatic and manual data policies shall achieve a short time-to-result. There are efforts to improve data policies. Data placement/allocation is vital to coping with the increasing amount of data processing in different data centers. A data allocation/placement policy decides which locations sub-sets of data are to be placed. In this paper, a novel approach copes with the bottleneck related to wide-area file transfers between data centers and large distributed data sets with high dimensionality. The model estimates similar data with a neural network on sparse and uncertain observations and then proceeds with the allocation process. The allocation process comprises evolutionary data allocation for finding near-optimal solutions and improves over 5% on network transfers for the given data centers.
Clustering analysis is a study which is of great value, and the large-scale government-data needed to be handled by cluster analysis is growing increasingly. Efficient analysis techniques of large-scale data need to b...
详细信息
ISBN:
(纸本)9781509035557
Clustering analysis is a study which is of great value, and the large-scale government-data needed to be handled by cluster analysis is growing increasingly. Efficient analysis techniques of large-scale data need to be adopted to handle the large-scale data. Traditional model of serial programming has serious scalability shortage, which don't satisfy the need of the large-scale government-data handling for computing and storage resources. distributedcomputing technology represented by the MapReduce has good scalability, and can greatly improve the execution efficiency of data-intensive algorithm, and give play to the computing power of compute cluster based on general hardware. Based on the background of "data platform for public petition", it aims to study how to combine the cluster analysis technology with the current massive government-data, extracting useful information from the mass characteristics hidden in the data through the cluster analysis technology, which can provide comprehensive analyse for system managers and decision makers. This paper focus on the study of combining basic distributed clustering algorithm and TF-IDF algorithm, developing the cases feature analysis module based on distributed clustering algorithm. Based on distributed clustering algorithm, according to the information of the cases, do clustering analysis of cases according to its characteristics, and then get several hidden information through serveral decisional result.
Many real-world wireless sensor network applications such as environmental monitoring, structural health monitoring, and smart grid can be formulated as a least-squares problem. In distributed Cyber-Physical System (C...
详细信息
ISBN:
(纸本)9781479988563
Many real-world wireless sensor network applications such as environmental monitoring, structural health monitoring, and smart grid can be formulated as a least-squares problem. In distributed Cyber-Physical System (CPS), each sensor node observes partial phenomena due to spatial and temporal restriction and is able to form only partial rows of least-squares. Traditionally, these partial measurements were gathered at a centralized location. However, with the increase in sensors and their measurements, aggregation is becoming challenging and infeasible. In this paper, we propose distributed randomized kaczmarz that performs in-network computation to solve least-squares over the network by avoiding costly communication. As a case study, we present a volcano monitoring application on a distributed CORE emulator and use real data from Mt. St. Helens to evaluate our proposed method.
暂无评论