| 一种基于跳跃hash的对象分布算法 |
| |
| 引用本文： | 聂世强,伍卫国,张兴军,蔡毅,徐志伟. 一种基于跳跃hash的对象分布算法[J]. 软件学报, 2017, 28(8): 1929-1939 |
| |
| 作者姓名： | 聂世强 伍卫国 张兴军 蔡毅 徐志伟 |
| |
| 作者单位： | 西安交通大学 电子与信息工程学院, 陕西 西安 710049,西安交通大学 电子与信息工程学院, 陕西 西安 710049,西安交通大学 电子与信息工程学院, 陕西 西安 710049,西安交通大学 电子与信息工程学院, 陕西 西安 710049,西安交通大学 电子与信息工程学院, 陕西 西安 710049 |
| |
| 基金项目： | 国家自然科学基金(91330117)；国家高技术研究发展计划(863)(2014AA01A302) |
| |
| 摘 要： | 如何有效地将海量数据分布到存储节点是存储系统首要解决的问题.本文提出的MJHAR（Matrix-basedJump Hash Algorithm for Replication data）对象分布算法简洁高效，支持权值和数据冗余机制.本算法创造性的将节点映射到二维矩阵，对象的分布、定位只需从矩阵的行内、行间计算目标节点的行号和列号即可.理论证明本算法满足公平性、自适应性、紧凑性、节点变化对象迁移量较小的特点，同时实验结果表明本算法的计算时间比一致性hash算法快40%，比跳跃hash算法快23%.极大降低了计算时间，并且比一致性hash算法对象分布更加均匀.
|
| 关 键 词： | 数据分布 对象存储系统 跳跃hash |
| 收稿时间： | 2016-06-19 |
| 修稿时间： | 2016-12-01 |
Object Placement Algorithm Based on Jump Hash |
| |
| NIE Shi-Qiang,WU Wei-Guo,ZHANG Xing-Jun,CAI Yi and XU Zhi-Wei. Object Placement Algorithm Based on Jump Hash[J]. Journal of Software, 2017, 28(8): 1929-1939 |
| |
| Authors: | NIE Shi-Qiang WU Wei-Guo ZHANG Xing-Jun CAI Yi XU Zhi-Wei |
| |
| Affiliation: | School of the Electronic and Information Engineering, Xi''an jiaotong University, Xi''an 710049, China,School of the Electronic and Information Engineering, Xi''an jiaotong University, Xi''an 710049, China,School of the Electronic and Information Engineering, Xi''an jiaotong University, Xi''an 710049, China,School of the Electronic and Information Engineering, Xi''an jiaotong University, Xi''an 710049, China and School of the Electronic and Information Engineering, Xi''an jiaotong University, Xi''an 710049, China |
| |
| Abstract: | Efficient and scalable object management is one major factor of the overall performance in the object-based storage system. This paper has proposed an algorithm MJHAR(Matrix-based Jump Hash Algorithm for Replication data) that could be applied to the heterogeneous storage system and support variable levels of the object replication, the algorithm could both runs very fast and moves the more minimizing data when storage nodes changes. The experiments show that the running time of MJHAR is less than that of the consistent hash by 40% and that of jump hash by 23%. The object distribution of MJHAR is more uniform than that of consistent hash algorithm. |
| |
| Keywords: | data distribution Object-Based Storage System jump consistent hash algorithm |
|
| 点击此处可从《软件学报》浏览原始摘要信息 |
|
点击此处可从《软件学报》下载全文 |
|
