2017 International Conference on High Performance Computing & Simulation (HPCS)最新文献_第4页

A Hybrid Parallel Algorithm for Solving Eeuler Equation Using Explicit RKDG Method Based on OpenFOAM 基于OpenFOAM的显式RKDG法求解Eeuler方程的混合并行算法

2017 International Conference on High Performance Computing & Simulation (HPCS) Pub Date : 2017-07-01 DOI: 10.1109/HPCS.2017.99

S. Ye, Xiaoguang Ren, Yuhua Tang, Liyang Xu, Hao Li, Chao Li, Yufei Lin

{"title":"A Hybrid Parallel Algorithm for Solving Eeuler Equation Using Explicit RKDG Method Based on OpenFOAM","authors":"S. Ye, Xiaoguang Ren, Yuhua Tang, Liyang Xu, Hao Li, Chao Li, Yufei Lin","doi":"10.1109/HPCS.2017.99","DOIUrl":"https://doi.org/10.1109/HPCS.2017.99","url":null,"abstract":"OpenFOAM is a framework of the open source C CFD toolbox for flexible engineering simulation, which uses finite volume method (FVM) in the discretization of partial differential equations (PDEs). The problem solving procedure in OpenFOAM consists in equations dicretization stage, equations solving stage and field limiting stage. The best parallelism is limited by the equation solving stage, which contains communications. Compared to FVM, discontinuous Galerkin (DG) method is a high-order discretization method, which can accelerate the convergence of the residuals over same mesh scale and has higher resolution of the flow. Based on OpenFOAM with DG method, the ratio of overhead in equations discretization stage increases, especially when solving Euler equations using an explicit method. The equations discretization stage has a better potential parallelism than the other two stages due to no existence of communication. In this paper, we will analysis the difference of time cost in these three stages between original OpenFOAM and OpenFOAM with DG method. By decoupling these three stages, a hybrid parallel algorithm for solving PDEs is proposed and implemented based on OpenFOAM with DG method. The experimental results show that the simulation time is reduced by 16%, and the relative speedup of the hybrid parallel algorithm is up to 2.88 compared to the original parallel algorithm with the same degree of parallelism.","PeriodicalId":115758,"journal":{"name":"2017 International Conference on High Performance Computing & Simulation (HPCS)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124791667","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Dynamic Resource Selection in Cloud Service Broker 云服务代理中的动态资源选择

2017 International Conference on High Performance Computing & Simulation (HPCS) Pub Date : 2017-07-01 DOI: 10.1109/HPCS.2017.43

G. Z. Santoso, Young-Woo Jung, Seong-Woo Seok, E. Carlini, Patrizio Dazzi, J. Altmann, John Violos, Jamie Marshall

引用次数: 11

A Topology-Adaptive Strategy for Graph Traversing 图遍历的拓扑自适应策略

2017 International Conference on High Performance Computing & Simulation (HPCS) Pub Date : 2017-07-01 DOI: 10.1109/HPCS.2017.60

Jia Meng, Liang Cao, Huashan Yu

{"title":"A Topology-Adaptive Strategy for Graph Traversing","authors":"Jia Meng, Liang Cao, Huashan Yu","doi":"10.1109/HPCS.2017.60","DOIUrl":"https://doi.org/10.1109/HPCS.2017.60","url":null,"abstract":"Graphs are a key form of Big Data. Although graph computing technology has been studied extensively in recent years, it remains a grand challenge to process large-scale graphs efficiently. Computation on a graph is to propagate and update the vertex values systematically. Both its complexity and parallelism are affected mainly by the algorithm's value propagating pattern. Efficient graph computing depends on techniques compatible with the algorithm's value propagating pattern. Graph traversing is a value propagating pattern used by representative graph applications. This paper presents an efficient value propagating framework for large-scale graph traversing applications. By partitioning the input graph based on the topology, it allows values for different source vertices to be propagated together, so as to reduce value propagating overhead. To improve the parallel efficiency of graph traversals, a novel task scheduling mechanism has been devised. The mechanism allows the framework to improve load balance without loss of locality. A prototype for the framework has been implemented. We evaluated the prototype with a set of typical real-world and synthetic graphs. By comparing with the owner-computing rule, experimental results show that this work has an overall speedup from 1.23 to 3.97. The speedup to Ligra is from 4.7 to 20.7.","PeriodicalId":115758,"journal":{"name":"2017 International Conference on High Performance Computing & Simulation (HPCS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128962474","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Minimizing Distribution and Data Loading Overheads in Parallel Training of DNN Acoustic Models with Frequent Parameter Averaging 基于频繁参数平均的DNN声学模型并行训练中的最小分布和数据负载开销

2017 International Conference on High Performance Computing & Simulation (HPCS) Pub Date : 2017-07-01 DOI: 10.1109/HPCS.2017.89

P. Rosciszewski, Jakub Kaliski

{"title":"Minimizing Distribution and Data Loading Overheads in Parallel Training of DNN Acoustic Models with Frequent Parameter Averaging","authors":"P. Rosciszewski, Jakub Kaliski","doi":"10.1109/HPCS.2017.89","DOIUrl":"https://doi.org/10.1109/HPCS.2017.89","url":null,"abstract":"In the paper we investigate the performance of parallel deep neural network training with parameter averaging for acoustic modeling in Kaldi, a popular automatic speech recognition toolkit. We describe experiments based on training a recurrent neural network with 4 layers of 800 LSTM hidden states on a 100-hour corpora of annotated Polish speech data. We propose a MPI-based modification of the training program which minimizes the overheads of both distributing training jobs and loading and preprocessing training data by using message passing and CPU/GPU computation overlapping. The impact of the proposed optimizations is greater for the more frequent neural network model averaging. To justify our efforts, we examine the influence of averaging frequency on the trained model efficiency. We plot learning curves based on the average log-probability per frame of correct paths for utterances in the validation set, as well as word error rates of test set decodings. Based on experiments with training on 2 workstations with 4 GPUs each we point that for the given network architecture, dataset and computing environment there is a certain range of averaging frequencies that are optimal for the model efficiency. For the selected averaging frequency of 600k frames per iteration the proposed optimizations reduce the training time by 54.9%.","PeriodicalId":115758,"journal":{"name":"2017 International Conference on High Performance Computing & Simulation (HPCS)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128820045","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 2

Parallelization of Large-Scale Drug-Protein Binding Experiments 大规模药物-蛋白质结合实验的并行化

2017 International Conference on High Performance Computing & Simulation (HPCS) Pub Date : 2017-07-01 DOI: 10.1109/HPCS.2017.39

Antonios Makris, D. Michail, Iraklis Varlamis, Chronis Dimitropoulos, K. Tserpes, G. Tsatsaronis, J. Haupt, M. Sawyer

{"title":"Parallelization of Large-Scale Drug-Protein Binding Experiments","authors":"Antonios Makris, D. Michail, Iraklis Varlamis, Chronis Dimitropoulos, K. Tserpes, G. Tsatsaronis, J. Haupt, M. Sawyer","doi":"10.1109/HPCS.2017.39","DOIUrl":"https://doi.org/10.1109/HPCS.2017.39","url":null,"abstract":"Drug polypharmacology or “drug promiscuity” refers to the ability of a drug to bind multiple proteins. Such studies have huge impact to the pharmaceutical industry, but in the same time require large investments on wet-lab experiments. The respective in-silico experiments have a significantly smaller cost and minimize the expenses for the subsequent lab experiments. However, the process of finding similar protein targets for an existing drug, passes through protein structural similarity and is a highly demanding in computational resources task. In this work, we propose several algorithms that port the protein similarity task to a parallel high-performance computing environment. The differences in size and complexity of the examined protein structures raise several issues in a naive parallelization process that significantly affect the overall time and required memory. We describe several optimizations for better memory and CPU balancing which achieve faster execution times. Experimental results, on a high-performance computing environment with 512 cores and 2048GB of memory, demonstrate the effectiveness of our approach which scales well to large amounts of protein pairs.","PeriodicalId":115758,"journal":{"name":"2017 International Conference on High Performance Computing & Simulation (HPCS)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120949198","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Data Exploration on Large Amount of Relational Data through Keyword Queries 基于关键字查询的海量关系数据挖掘

2017 International Conference on High Performance Computing & Simulation (HPCS) Pub Date : 2017-07-01 DOI: 10.1109/HPCS.2017.21

D. Beneventano, F. Guerra, Yannis Velegrakis

{"title":"Data Exploration on Large Amount of Relational Data through Keyword Queries","authors":"D. Beneventano, F. Guerra, Yannis Velegrakis","doi":"10.1109/HPCS.2017.21","DOIUrl":"https://doi.org/10.1109/HPCS.2017.21","url":null,"abstract":"The paper describes a new approach for querying relational databases through keyword search by exploting Information Retrieval (IR) techniques. When users do not know the structures and the content, keyword search becomes the only efficient and effective solution for allowing people exploring a relational database. The approach is based on a unified view of the database relations (performed through the full disjunction operator), where its composing tuples will be considered as documents to be indexed and searched by means of an IR search engine. Moreover, as it happens in relational databases, the system can merge the data stored in different documents for providing a complete answer to the user. In particular, two documents can be joined because either their tuples in the original database share some Primary Key or, always in the original database, some tuple is connected by a Primary / Foreign Key Relation. Our preliminary proposal, the description of the tabular data structure for storing and retrieving the possible connections among the documents and a metrics for scoring the results are introduced in the paper.","PeriodicalId":115758,"journal":{"name":"2017 International Conference on High Performance Computing & Simulation (HPCS)","volume":"124 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133304252","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 2

Big-Data in Climate Change Models — A Novel Approach with Hadoop MapReduce 气候变化模型中的大数据——Hadoop MapReduce的一种新方法

2017 International Conference on High Performance Computing & Simulation (HPCS) Pub Date : 2017-07-01 DOI: 10.1109/HPCS.2017.17

J. C. Loaiza, G. Giuliani, G. Fiameni

引用次数: 2

Integrating Heterogeneous Weather-Sensors Data into a Smart-City App 将异构天气传感器数据集成到智能城市应用程序中

2017 International Conference on High Performance Computing & Simulation (HPCS) Pub Date : 2017-07-01 DOI: 10.1109/HPCS.2017.33

A. Quarati, A. Clematis, Luca Roverelli, Gabriele Zereik, D. D'Agostino, G. Mosca, M. Masnata

{"title":"Integrating Heterogeneous Weather-Sensors Data into a Smart-City App","authors":"A. Quarati, A. Clematis, Luca Roverelli, Gabriele Zereik, D. D'Agostino, G. Mosca, M. Masnata","doi":"10.1109/HPCS.2017.33","DOIUrl":"https://doi.org/10.1109/HPCS.2017.33","url":null,"abstract":"Current weather information is one of the facilities supplied to the users by most travel and mobility systems. Most of them relies on one provider, who can deliver a more or less satisfactory coverage on different geographical areas. In this paper, we present the approach and discuss the rationale that drove the design and development of a mashup service for providing meteorological information within the research project TCUBE (“Transport Territory and Tourism“). TCUBE aims at the creation of a platform of services for travel and mobility in urban territory, with initial focus on Genoa metropolitan area. The goal of the project is to study, develop and validate technologies and solutions for the implementation, deployment, and management of advanced information services for citizens and visitors based on enabling technologies and methodologies such as: Open Data; crowd sourcing and social sharing; sensor infrastructures and services based on spatial data. The mashup service presented herein is able to exploit and integrate weather sensor data provided by the free contributions of citizen scientists' Personal Weather Stations belonging to heterogeneous Weather Networks. Design and technical details of our approach are supplied, thus to make it replicable in other similar urban contexts.","PeriodicalId":115758,"journal":{"name":"2017 International Conference on High Performance Computing & Simulation (HPCS)","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131766658","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 11

Design of Cache Backend Using Remote Memory for Network File System 基于远程内存的网络文件系统缓存后端设计

2017 International Conference on High Performance Computing & Simulation (HPCS) Pub Date : 2017-07-01 DOI: 10.1109/HPCS.2017.131

Eunji Lim, Shinyoung Ahn, Youngho Kim, Gyuil Cha, Wan-Sik Choi

引用次数: 3

When is the Right Time to Start the Fault Tolerance Protection? 什么时候启动容错保护是合适的?

2017 International Conference on High Performance Computing & Simulation (HPCS) Pub Date : 2017-07-01 DOI: 10.1109/HPCS.2017.70

Jorge Villamayor, Dolores Rexachs, E. Luque

{"title":"When is the Right Time to Start the Fault Tolerance Protection?","authors":"Jorge Villamayor, Dolores Rexachs, E. Luque","doi":"10.1109/HPCS.2017.70","DOIUrl":"https://doi.org/10.1109/HPCS.2017.70","url":null,"abstract":"In High Performance Computing, Fault Tolerance (FT) becomes a primary concern due to the constant growing and continuous aging of hardware components, which rise failures probability. Failures produce performance degradation to the environment and affect significantly users expected execution time. Rollback-Recovery protocols represent a fundamental component to protect and restore users parallel application execution, although this protection comes with an overhead. This paper proposes a First Protection Point model, which determines the starting point to introduce FT protection gaining benefits in terms of total execution time including failures. A characterization of Rollback-Recovery protocols applied on parallel applications is performed, to obtain key factors for the model design. This model can help users determine which checkpoints can be removed from the application execution when they are used for FT protection purposes, reducing the overhead and at the same time keeping high availability. An analytic model evaluation is developed to show the inflexion point where FT protection starts to provide benefits for users. Finally, three experimental environments are setup, using two private clusters and a public cluster configured in a well-known cloud Amazon EC2. A coordinated checkpoint facility is applied on NAS benchmark applications such as: CG, BT and LU to evaluate the proposed model, obtaining overhead impact reduction for provided Fault Tolerance.","PeriodicalId":115758,"journal":{"name":"2017 International Conference on High Performance Computing & Simulation (HPCS)","volume":"112 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133675071","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0