FAULT-TOLERANT DATA SHARING FOR HIGH- LEVEL GRID PROGRAMMING A HIERARCHICAL STORAGE ARCHITE.pdf

FAULT-TOLERANT DATA SHARING FOR HIGH- LEVEL GRID PROGRAMMING A HIERARCHICAL STORAGE ARCHITE.pdf

  1. 1、本文档共12页,可阅读全部内容。
  2. 2、原创力文档(book118)网站文档一经付费(服务费),不意味着购买了该文档的版权,仅供个人/单位学习、研究之用,不得用于商业用途,未经授权,严禁复制、发行、汇编、翻译或者网络传播等,侵权必究。
  3. 3、本站所有内容均由合作方或网友上传,本站不对文档的完整性、权威性及其观点立场正确性做任何保证或承诺!文档内容仅供研究参考,付费前请自行鉴别。如您付费,意味着您自己接受本站规则且自行承担风险,本站不退款、不进行额外附加服务;查看《如何避免下载的几个坑》。如果您已付费下载过本站文档,您可以点击 这里二次下载
  4. 4、如文档侵犯商业秘密、侵犯著作权、侵犯人身权等,请点击“版权申诉”(推荐),也可以打举报电话:400-050-0827(电话支持时间:9:00-18:30)。
查看更多
FAULT-TOLERANT DATA SHARING FOR HIGH- LEVEL GRID PROGRAMMING A HIERARCHICAL STORAGE ARCHITE

FAULT-TOLERANT DATA SHARING FOR HIGH- LEVEL GRID PROGRAMMING: A HIERARCHICAL STORAGE ARCHITECTURE Marco Aldinucci, Marco Danelutto Computer Science Department, University of Pisa Largo Bruno Pontecorvo 3, 56127 Pisa, Italy aldinuc@di.unipi.it Gabriel Antoniu, Mathieu Jan INRIA Rennes Campus de Beaulieu, 35042 Rennes, France gabriel.antoniu@irisa.fr, mathieu.jan@irisa.fr Abstract Enabling high-level programming models on grids is today a major challenge. A way to achieve this goal relies on the use of environments able to transparently and automatically provide adequate support for low-level, grid-specific issues (fault-tolerance, scalability, etc.). This paper discusses the above approach when applied to grid data management. As a case study, we propose a 2-tier software architecture that supports transparent, fault-tolerant, grid-level data sharing in the ASSIST programming environment (University of Pisa), based on the JuxMem grid data sharing service (INRIA Rennes). Keywords: Grid, shared memory, High-level programming, memory hierarchy, P2P. 21. Introduction Grid computing has emerged as an attempt to provide users with the illusion of an infinitely powerful, easy-to-use computer, which can solve very complex problems. This very appealing illusion is to be provided (1) by relying on the aggregated power of standard (so, inexpensive), geographically distributed resources owned by multiple organizations; (2) by hiding as much as possible the complexity of the distributed infrastructure to users. However, the current status in most software grid infrastructures available today is rather far away from this vision. When designing programs able to run on such large-scale platforms, programmers often need to explicitly take into account resource heterogeneity, as well as the unreliability of the distributed infrastructure. In this context, the grid community converges towards a consensus about the need for a high-level programming model, whereas most of the grid-specif

文档评论(0)

l215322 + 关注
实名认证
内容提供者

该用户很懒,什么也没介绍

1亿VIP精品文档

相关文档