摘要
提出一种基于喷泉码、网络编码思想的数据备份方案。该方案基于喷泉码、网络编码的思想,并且具有纠删码的特性。在文件存储时,将文件块进行编码,产生很多编码块,对原始文件块和编码块进行挑选存储,最后进行存储的是原始块与编码块的混合,这样文件存储空间会变大。数据读取时,只需经过简单编码计算即可,一般都是线性运算。该方案利用空间和计算资源来换取数据的可靠性,在某种应用场景下有更好的鲁棒性。
Proposes a data backup scheme which is based on fountain coding and network coding. This scheme is based on the idea of fountain coding and network coding, and it has the characteristics of erasure codes. When data are stored, the file blocks will be encoded then produce a lot of encoding blocks, after that the original block and file encoding block will be selected to storage and the store data is mixed with the original block and the encoding block so the file storage space becomes larger. When data are read, the original data can be recovered by a simple calculation which is generally linear operations. This scheme obtains the reliability of data by exploiting the space and calculation, it has a better robustness in some application scenes.
基金
国家自然科学基金项目(No.61272013)
广东省现代信息服务业发展专项资金项目(No.GDEID2011IS022)
广东省省部产学研合作专项资金项目(No.2013B090500007)
关键词
喷泉码
网络编码
备份
鲁棒性
Fountain Code
Encoding
Backup
Robustness
