116 Slices
Medium 9781601323262

Processing NOAA Observation Data over Hybrid Computer Systems for Comparative Climate Change Analysis

Hamid R. Arabnia, Lou D'Alotto, Hiroshi Ishii, Minoru Ito, Kazuki Joe, Hiroaki Nishikawa, Georgios Sirakoulis, William Spataro, Giuseppe A. Trunfio, George A. Gravvanis, George Jandieri, Ashu M. G. Solo, Fernando G. Tinetti CSREA Press PDF

10

Int'l Conf. Par. and Dist. Proc. Tech. and Appl. | PDPTA'14 |

Processing NOAA Observation Data over Hybrid

Computer Systems for Comparative Climate Change

Analysis

Xuan Shi1,, Dali Wang 2

Department of Geosciences, University of Arkansas, Fayetteville, AR 72701, USA

2

Environmental Science Division, Oak Ridge National Lab Oak Ridge TN 37831, USA

1

Abstract - With the rapid development of weather monitoring system, numerous observational data are available. For example, NOAA provides Global Surface Summary of Day

(GSOD) data that incorporates daily weather measurements from over 9000 weather stations around the world. In this paper, a comprehensive workflow and methodology is presented to elaborate how to transform GSOD data into a new and useful format so as to generate interpolated product of daily, monthly and annual mean surface temperature datasets by using advanced computation platforms. The quality of this gridded, high resolution (at ¼ degree) daily product is further examined by comparing to an existing global climate dataset. A preliminary comparison on global surface temperature shows a consistent agreement between these two datasets, with the major differences located in a few regions. The interpolated GSOD data products are supplementary to existing datasets by providing new gridded, high resolution observation-based daily temperature information over three decades (1982-2011), which are very useful for decadal climate change researches.

See All Chapters
Medium 9781601323262

Emotion Estimation of Comments on Web News by SVM and Naive Bayes Based Classifiers

Hamid R. Arabnia, Lou D'Alotto, Hiroshi Ishii, Minoru Ito, Kazuki Joe, Hiroaki Nishikawa, Georgios Sirakoulis, William Spataro, Giuseppe A. Trunfio, George A. Gravvanis, George Jandieri, Ashu M. G. Solo, Fernando G. Tinetti CSREA Press PDF

Int'l Conf. Par. and Dist. Proc. Tech. and Appl. | PDPTA'14 |

189

Emotion Estimation of Comments on Web News by SVM and

Naive Bayes Based Classifiers

Yasuhiro Tajima and Genichiro Kikui

Department of Systems Engineering, Okayama Prefectural University,

111, Kuboki, Soja, Okayama 719-1197, Japan

Abstract— Social communication tools such as Twitter or

Facebook spread the web service ability. Using their APIs, we can gather many users’ comments easily. Such comments are usually short sentences but they also have many emotional comments. In this paper, we propose emotion estimation methods for multilabeled short comments of web news. Our methods can be applied to sentiment analysis and opinion mining. At first, we show the performance evaluation of a naive Bayes classifier and an SVM classifier.

Then, we propose two improved methods. The first is an improved naive Bayes method which classifies each emotion label into two opposite emotions and uses their weights. We call this the weighting method. The second method consists of two stages of classifiers. The first stage distinguishes these oppositely classes, and the second stage selects one emotion from the opposite emotions. From our evaluation, we conclude that the weighting method is better among the naive Bayes classifiers and its performance is as good as

See All Chapters
Medium 9781601323262

Viability of the Parallel Prefix Algorithm for Sequence Alignment on Massively Parallel GPUs

Hamid R. Arabnia, Lou D'Alotto, Hiroshi Ishii, Minoru Ito, Kazuki Joe, Hiroaki Nishikawa, Georgios Sirakoulis, William Spataro, Giuseppe A. Trunfio, George A. Gravvanis, George Jandieri, Ashu M. G. Solo, Fernando G. Tinetti CSREA Press PDF

Viability of the Parallel Prefix Algorithm for Sequence Alignment on Massively Parallel GPUs

A. Christian Dicker1 , B. John Sibandze1 , C. Samuel Kelly1 , and D. Jens Mache1

1 Computer Science, Lewis & Clark College, Portland, Oregon, USA

Abstract— In this paper, we compare two different methods for parallelizing the Needleman–Wunsch dynamic programming algorithm for finding the optimal alignment of two sequences: (1) computing the elements of each diagonal of the table in parallel and (2) computing the elements of each row in parallel using the parallel prefix algorithm. In

2003, the latter algorithm was shown to decrease communication between processors and provide a more uniform work distribution [1]. With the increasing prevalence of general purpose programming on graphics processing units (GPUs), there is a need to reassess the viability of the parallel prefixbased algorithm. We discuss our implementation of both algorithms on a massively parallel GPU and compare the runtimes by thread count as well as by sequence size. We

See All Chapters
Medium 9781601323262

Linux Software RAID Level 0 Technique for High Performance Computing by using PCI-Express based SSD

Hamid R. Arabnia, Lou D'Alotto, Hiroshi Ishii, Minoru Ito, Kazuki Joe, Hiroaki Nishikawa, Georgios Sirakoulis, William Spataro, Giuseppe A. Trunfio, George A. Gravvanis, George Jandieri, Ashu M. G. Solo, Fernando G. Tinetti CSREA Press PDF

Int'l Conf. Par. and Dist. Proc. Tech. and Appl. | PDPTA'14 |

107

Linux Software RAID Level 0 Technique for

High Performance Computing by using

PCI-Express based SSD

Jae Gi Son, Taegyeong Kim, Kuk Jin Jang, *Hyedong Jung

Department of Industrial Convergence, Korea Electronics Technology Institute, Korea jgson@keti.re.kr, taegyeong@keti.re.kr, kjang@keti.re.kr, *hudson@keti.re.kr

Abstract—The Linux-based legacy server systems are configured and used with software RAID to improve the performance of the disk I/O. Server systems requiring high performance prefer a special SSD that connects directly with the PCI-express bus to the SATA interface. However, the problem is that the current Linux kernel and software

RAID are difficult to optimize the high-performance SSD based on PCI-Express because it is designed to be optimized for the conventional hard disk. Therefore, we propose the efficient method using re-combination and remapping techniques to improve the performance of software RAID level-0 provided on the Linux kernel level.

See All Chapters
Medium 9781601323262

Maximizing Hardware Performance via Non-blocking Collective Communication for All Pairs Shortest Paths Computation on Heterogeneous Multi-core Processors

Hamid R. Arabnia, Lou D'Alotto, Hiroshi Ishii, Minoru Ito, Kazuki Joe, Hiroaki Nishikawa, Georgios Sirakoulis, William Spataro, Giuseppe A. Trunfio, George A. Gravvanis, George Jandieri, Ashu M. G. Solo, Fernando G. Tinetti CSREA Press PDF

66

Int'l Conf. Par. and Dist. Proc. Tech. and Appl. | PDPTA'14 |

Maximizing Hardware Performance via Non-blocking

Collective Communication for All Pairs Shortest Paths

Computation on Heterogeneous Multi-core Processors

Eduardo A. Colmenares1, Yu Zhuang1

1

Department of Computer Science, Texas Tech University, Lubbock, Texas, USA

Abstract - Dynamic data sharing among cores during computation in a multi-core architectural environment has been recognized as one of the factors that add to the cost of the total execution time. One way of reducing the impact of the latency generated by intense data sharing is through communication hiding.

The idea behind communication hiding is to create opportunities in the computation algorithms that can make use of system and/or hardware resources that allow the processors to engage in useful work while the sharing of information is taking place. One of the possible ways to achieve overlapping of communication and computation is by making use of non-blocking resources. In this paper, we present an algorithmically restructured solution for the All

See All Chapters

See All Slices