Web流语义感知的改进队列管理算法  

Improved Queue Management Algorithm for Semantic Aware of Web Flows

在线阅读下载全文

作  者:赵东平[1] 张德运[1] 高磊[1] 郑卫斌[1] 

机构地区:[1]西安交通大学电子与信息工程学院,西安710049

出  处:《西安交通大学学报》2006年第10期1047-1051,共5页Journal of Xi'an Jiaotong University

基  金:国家自然科学基金资助项目(60403028)

摘  要:对Web流的传输特性进行了分析,并在现有尾部丢弃和随机早期检测算法的基础上提出了基于流分组语义感知的改进队列管理算法.该算法首先保存有限时间滑动窗口内的活动流特征,对到达的每个分组根据其分组语义分别赋予不同的丢弃代价,而在检测到拥塞且需要丢弃分组时,优先丢弃代价较小的,避免丢弃关键的,从而减少分组丢失对Web传输造成的负面影响.仿真结果分析表明,使用所提算法可平均提高8%的有效吞吐率,平均缩短10%的响应时间,从整体上改善了Web流的传输性能.The transfer characteristics of Web flows are analyzed. On the basis of existing taildrop and random early detection algorithms, an improved queue management algorithm based on packet semantic-aware is proposed. Firstly all the active flows' characteristics are saved within a limited time sliding window, then various discard cost is assigned to each arrival packet according to its semantic-aware respectively. The packet with less cost is discarded first whenever the congestion is detected and packet needs to be discarded. Thereby it avoids the discard of the key packets and decreases the negative influence on Web transmission due to packet loss. The simulation shows that the proposed algorithm increases averagely the effective throughput rate by more than 8% and reduces averagely the response time by 10%,hence the transmission performance of Web flows is improved wholly.

关 键 词:流分组 队列管理 语义感知 丢弃代价 

分 类 号:TP393[自动化与计算机技术—计算机应用技术]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象