eScience 2012 has accepted 48 papers (and respective presentations) from the 160 submissions, based on peer reviews by a committee of scientific and technical domain experts. The papers will be presented in groups of ...
eScience 2012 has accepted 48 papers (and respective presentations) from the 160 submissions, based on peer reviews by a committee of scientific and technical domain experts. The papers will be presented in groups of three parallel sessions, covering: the infrastructure needed to support eScience, such as workflows, the process of eScience, data provenance, middleware, and semantics; and applications such as ecology, biology, astronomy, engineering, image processing, and social science. These papers are published in the conference proceedings. The diverse program provides a snapshot of the huge volume of scientific expertise and technological innovations that exist in the field today. Additionally, eScience 2012 will include about 25 posters covering early results and works-in-progress, based on the peer-reviews of about 35 submissions. 2-page papers for each of these accepted posters appear on the conference web site. The eScience 2012conference also includes 5 workshops: Extending High-Performance computing Beyond its Traditional User Communities (XHPC 2012), 2ndinternational Workshop on Analyzing and Improving Collaborative eScience with Social Networks (eSoN 12), Maintainable Software Practices in e-Science (SoftwarePractice 2012), eScience Meets the Instrument, Collaborative research using eScience infrastructure and high speed networks (NECS 2012); and one tutorial: Big Data Processing: Lessons from Industry and Applications in Science. Papers from the workshops were peer-reviewed and are also including in these proceedings.
cloud computing through the basic idea is to construct a large-scale cluster system based data center, the cluster of resources (such as hardware, development platform for virtual form) to provide users with a resourc...
详细信息
cloud computing through the basic idea is to construct a large-scale cluster system based data center, the cluster of resources (such as hardware, development platform for virtual form) to provide users with a resource pool. In this paper we in cloud computing and data processing techniques work, introduced emphatically based on cloud computingdistributedparallel programming model, a distributed task scheduling and cloud computing platform adaptive resource scheduling and management, and the difficulties and challenges are to accept;at the same time, but also to the future potential breakthrough direction the prospect of.
The functions of data exchange provide a series of methods, transfer data from a storage system to another storage system. HDFS is a distributed File System realized by Hadoop, which has the character of high fault-to...
详细信息
The functions of data exchange provide a series of methods, transfer data from a storage system to another storage system. HDFS is a distributed File System realized by Hadoop, which has the character of high fault-tolerance, at the same time it provides a high transfer rate to access the data of applications, and is suitable for those applications with large data set. Traditionally, the large data is stored in FTP servers or SQL databases. We use Hadoop distributed framework for large-scale data calculation, which will certainly need to transfer data from FTP servers or SQL databases to HDFS. This paper mainly discussed the problem of parallel data exchange between SQL database and HDFS, introduced the performance of Hadoop data exchange functions: DBInputFormat/ DBOutputFormat, and put forward some strategies to improve the performance.
暂无评论