Abstract: Today scientific applications on distributed computing environment have huge data transfer which continues to increase drastically every year. This implies a major necessity to move huge amounts of data from original data site to target site on the whole computation cycle, which brings with it the problem of efficient and reliable data placement. The current approach to solve this problem of data placement is either doing it manually, or employing simple scripts which do not have any automation or fault tolerance capabilities. The goal is to make data placement activities robust and efficient. It will be queued, scheduled, monitored, managed, and even check-pointed. The data placement activities should be treated differently from computational jobs, since they may have different semantics and different characteristics. The method for data placement is tested.

Key words: data placement, reliable transfer, distributed computing

摘要： 数据放置活动在以计算为主的分布式计算中被看作是次要的任务。文章将数据放置工作与计算工作区别对待，定义分布式计算过程中的数据放置模型，给出数据放置协议和可靠的数据传输算法，使得数据放置活动在分布式计算中像计算工作一样被排列、调度、监控、管理以及检测。对数据放置方法进行了系统的评价，取得了较好的效果。

关键词: 数据放置, 可靠传输, 分布式计算

CLC Number: 

• TP301.6