欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
FIB表急剧增长是互联网高速发展面临的重要问题之一，FIB表分解存储能有效解决该问题．现有的SPAI，技术将FIB表较均匀地分解存储在不同线卡，但仍然存在较多的表项冗余存储现象，并且实现复杂．对此设计了一种新型的转发表分解存储模型（Decomposed StorageofFIB，DSF），它依据IP前缀的前若干bit位实现线卡对转发表的分解存储，并只带来极少的冗余存储．DSF的改进方案——EDSF，更可使各线卡非常均衡地完成分解存储．提出的分解存储模型缓解了FIB表项急剧增长问题的解决压力，同时大大节省了硬件资源．对于线卡数量更多的可扩展路由器尤为适合．通过对当前运营的路由表的分解存储实验研究及与其它方案的比较，验证了文中模型良好的存储性能．
英文摘要：
The fast increasing of FIB has put unprecedented pressure to the Internet routers. The decomposed storage of FIB has theretofore been widely suggested, which explores the balance of storage overhead over routers＇ line cards. SPAL can achieve the decomposed storage of FIB, but its implement is very complex, and there exists many redundancy storage of FIB. In this paper, we propose a novel decomposed storage model of FIB, DSF（Decomposed Storage of FIB）, which can reduce the unnecessary redundancy of FIB storage across line cards. Moreover, we discuss an enhanced mechanism： EDSF. The real-FIB experiments show that EDSF can further minimize the diversity of line cards＇ storage. The design and the implementation of our proto type is also validated in detail showing that the proposed architecture is compatible with the existing commercial coretouters.
同期刊论文项目
同项目期刊论文
位置：