The proceedings contain 24 papers. The topics discussed include: realising parallel and distributedsimulation in industry: a roadmap;distribution of random streams in stochastic models in the age of multi-core and ma...
ISBN:
(纸本)9781457713668
The proceedings contain 24 papers. The topics discussed include: realising parallel and distributedsimulation in industry: a roadmap;distribution of random streams in stochastic models in the age of multi-core and manycore processors;parallel discrete event N-body dynamics;modeling billion-node torus networks using massively parallel discrete-event simulation;empirical study on entity interaction graph of large-scale parallelsimulations;collaborative interest management for peer-to-peer networked virtual environment;dead reckoning-based update scheduling against message loss for improving consistency in DVEs;trading computation time for synchronization time in spatial distributedsimulation;framework for simulation of hybrid systems: interoperation of discrete event and continuous simulators using HLA/RTI;an analysis of the cost of validating semantic composability;and fairness verification of BOM-based composed models using petri nets.
Predicting and analyzing runtime performance characteristics is a vital step in the development process of parallel discrete event simulations. For instance, model developers need to identify and eliminate performance...
详细信息
Networks for the smart grids are characterized by millions of sensor nodes exchanging information about the status of the grid. This information exchange must be realized reliably and efficiently due to the mission cr...
详细信息
This book represents the combined peer-reviewed proceedings of the Fifth International Symposium on Intelligent distributed Computing -- IDC 2011 and of the Third International workshop on Multi-Agent Systems Technolo...
ISBN:
(纸本)9783642240126
This book represents the combined peer-reviewed proceedings of the Fifth International Symposium on Intelligent distributed Computing -- IDC 2011 and of the Third International workshop on Multi-Agent Systems Technology and Semantics -- MASTS 2011. Both events were held in Delft, The Netherlands during October 5-7, 2011. The 33 contributions published in this book address many topics related to theory and applications of intelligent distributed computing and multi-agent systems, including: adaptive and autonomous distributed systems, agent programming, ambient assisted living systems, business process modeling and verification, cloud computing, coalition formation, decision support systems, distributed optimization and constraint satisfaction, gesture recognition, intelligent energy management in WSNs, intelligent logistics, machine learning, mobile agents, parallel and distributed computational intelligence, parallel evolutionary computing, trust metrics and security, scheduling in distributed heterogenous computing environments, semantic Web service composition, social simulation, and software agents for WSNs.
Particle swarm optimization (PSO) is a powerful technique for computer aided prediction of proteins' three-dimensional structure. In this work, employing an all-atom force field we demonstrate the efficiency of th...
详细信息
Advancements in massively parallel sampling of the conformational space of biomolecules enables, for example, protein structure prediction, in-silico drug development and cell signaling. Despite the existence of highl...
详细信息
ISBN:
(纸本)9783866446939
Advancements in massively parallel sampling of the conformational space of biomolecules enables, for example, protein structure prediction, in-silico drug development and cell signaling. Despite the existence of highly distributed protein simulation architectures like POEM@HOME, there was no abundant computational resource both strong in serial strength and in parallel sampling. In this study we investigate the optimization of our N-body Lennard Jones force field for the efficient Monte-Carlo sampling of small to medium-size biomolecules on massively parallel architectures, like modern GPUs. We benchmark both NVIDIA and AMD GPU chipsets in the OpenCL framework in comparison to CPU architectures. The N-body interactions are broken down into small local grids, which fit into the local GPU caches to permit simultaneous evaluation. Using the N-body term we accelerate the Lennard-Jones and Clash-Potential of the complete freeenergy PFF02 [1] shown to fold a multitude of different protein-folds and implement a modified structurebased Lennard Jones force field. We proof the applicability of our novel force field by reversible foldingsimulations of a three-helix protein using this Go-potential from completely unfolded structures.
The proceedings contain 36 papers and 2 Power Point presentations. The topics discussed include: qualifying reliability of solid-state storage from multiple aspects;a program context-aware data separation technique fo...
ISBN:
(纸本)9781457704284
The proceedings contain 36 papers and 2 Power Point presentations. The topics discussed include: qualifying reliability of solid-state storage from multiple aspects;a program context-aware data separation technique for reducing garbage collection overhead in NAND flash memory;towards simulation of parallel file system scheduling algorithms with PFSsim;scalable distributed directory implementation on orange file system;hierarchical RAID (HRAID): organization, operation, reliability, and performance;a statistical evaluation of the impact of parameter selection on storage system benchmarks;performance models of flash-based solid-state drives for real workloads;sampling-based garbage collection metadata management for flash-based storage;data allocation strategies for the management of quality of service in virtualised storage systems;and RAID6L: a log-assisted RAID6 storage architecture with improved write performance.
The proceedings contain 14 papers. The topics discussed include: optimal multi-image processing streaming framework on parallel heterogeneous systems;parallel gradient domain processing of massive images;distributed O...
ISBN:
(纸本)9783905674323
The proceedings contain 14 papers. The topics discussed include: optimal multi-image processing streaming framework on parallel heterogeneous systems;parallel gradient domain processing of massive images;distributed OpenGL rendering in network bandwidth constrained environments;revisiting parallel rendering for shared memory machines;cross-segment load balancing in parallel rendering;load balancing utilizing data redundancy in distributed volume rendering;real-time ray tracer for visualizing massive models on a cluster;interactive particle tracing in time-varying tetrahedral grids;efficient I/O for parallel visualization;parallel computational steering and analysis for HPC applications using a ParaView interface and the HDF5 DSM virtual file driver;parallel in situ coupling of simulation with a fully featured visualization system;and a preview and exploratory technique for large-scale scientific simulations.
We consider the problem of optimizing the execution of dataintensive scientific workflows in the Cloud. We address the problem under the following scenario. The tasks of the workflows communicate through files;the out...
详细信息
ISBN:
(纸本)9781450307048
We consider the problem of optimizing the execution of dataintensive scientific workflows in the Cloud. We address the problem under the following scenario. The tasks of the workflows communicate through files;the output of a task is used by another task as an input file and if these tasks are assigned on different execution sites, a file transfer is necessary. The output files are to be stored at a site. Each execution site is to be assigned a certain percentage of the files and tasks. These percentages, called target weights, are pre-determined and reflect either user preferences or the storage capacity and computing power of the sites. The aim is to place the data files into and assign the tasks to the execution sites so as to reduce the cost associated with the file transfers, while complying with the target weights. To do this, we model the workflow as a hypergraph and with a hypergraph-partitioning-based formulation, we propose a heuristic which generates data placement and task assignment schemes simultaneously. We report simulation results on a number of real-life and synthetically generated scientific workflows. Our results show that the proposed heuristic is fast, and can find mappings and assignments which reduce file transfers, while respecting the target weights. Copyright 2011 ACM.
In this paper, we consider distributed task allocation in distributed systems. In order to improve the performance, using load information of nodes is promising. In reality, however, it is common that load information...
详细信息
ISBN:
(纸本)9780769543499
In this paper, we consider distributed task allocation in distributed systems. In order to improve the performance, using load information of nodes is promising. In reality, however, it is common that load information is not up to date when it is used. Load information may be updated only periodically, or the communication delay for a task to move to other node makes load information out of date. The performance of task allocation schemes is significantly damaged if the load information is old. In order to tackle this problem, we introduce the nearest neighbor approach which is originally proposed for parallel computing. Restricting the communication to between neighboring nodes, the effect of communication delay is suppressed. simulation results showed that our proposed scheme suppresses the effect of old load information and shows better performance than a previously proposed scheme.
暂无评论