| 一种基于FCBF的流信息抽样测量框架及算法* |
| |
| 引用本文： | 张峰,谭兴晔,雷振明.一种基于FCBF的流信息抽样测量框架及算法*[J].计算机应用研究,2005,22(6):38-41. |
| |
| 作者姓名： | 张峰 谭兴晔 雷振明 |
| |
| 作者单位： | 北京邮电大学,ATM技术研究中心,北京,100876 |
| |
| 基金项目： | 国家重大自然科学基金资助项目(69896240) |
| |
| 摘 要： | 基于FCBF的高效流信息抽样测量框架不仅可以抽样测量三类流参数，而且存储开销小，只需1～3MB字节左右的存储空间；同时还可以做到几乎零概率的流信息识别统计误差。分析结果表明，该算法可以支持远高于OC48的链路速率，甚至可达OC192或更高；适合于将来高速链路上细粒度的流信息抽样测量。
|
| 关 键 词： | BF FCBF 流信息 抽样测量 开销 |
| 文章编号： | 1001-3695(2005)06-0038-04 |
| 修稿时间： | 2004年6月7日 |
Architecture and Algorithm for Flow Information Sampling Measurement Based on FCBF |
| |
| ZHANG Feng,TAN Xing-ye,LEI Zhen-ming.Architecture and Algorithm for Flow Information Sampling Measurement Based on FCBF[J].Application Research of Computers,2005,22(6):38-41. |
| |
| Authors: | ZHANG Feng TAN Xing-ye LEI Zhen-ming |
| |
| Affiliation: | (ATM Research & Develop Center, Beijing University of Posts & Telecommunications, Beijing 100876, China) |
| |
| Abstract: | The architecture and algorithm which can sampling measure three types of flow information are proved that they are feasible to be implemented in hardware with low memory space required and little computation overhead, and can acquire the effect of almost zero probability on identifying and counting flow information working on a link faster than OC-48, even reaching OC-192 or even higher. So the architecture and algorithm adaptive to be implemented in current high-speed network link interface are suitable to measure flow distribution information with high resolution and fine granularity. |
| |
| Keywords: | BF(Bloom Filter) FCBF(Fragment Counting Bloom Filter) Flow Information Sampling Measurement Overhead |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
| 点击此处可从《计算机应用研究》浏览原始摘要信息 |
|
点击此处可从《计算机应用研究》下载全文 |
