Scalable, Structured Data Placement over P2P Storage Utilities

Zhang, Zheng ; Mahalingam, Mallik ; Xu, Zhichen ; Tang, Wenting

HP Development Company

关键词: P2P storage utility;    metadata placement;    structured data;    CAN;    namespace;    tree;   

RP-ID  :  HPL-2002-40

学科分类：计算机科学（综合）

美国|英语

来源: HP Labs

【 摘 要 】

Current peer-to-peer storage utilities offer a convenient flat storage space, delegating the organization and presentation of data to upper layers. In reality, both applications and users typically organize data in a structured form. One such popular structure is hierarchical namespace as employed in a file system. A naive approach such as hashing the pathname of file system not only ignores locality in important operations such as file/directory lookup, but also results in uncontrollable, massive object relocations when rename on path component occurs. In this paper, we investigate policies and strategies that map the hierarchical namespace onto the flat storage space of P2P systems. We found that, in general, there exists a tradeoff between lookup performance and balanced storage utilization, and attempts to balance these two requirements calls for intelligent placement decision. We show that simple heuristics are effective in achieving significant performance benefit with negligible overhead. In addition, combining some of the heuristics and carefully setting the parameters can significantly reduce the lookup cost while keeping the impact on storage utilization minimal. These algorithms are robust and generic, capable of handling data layout to capture access locality. 12 Pages