DTN中基于消息综合属性的缓存管理策略  

Buffer Management Strategy Based on Message Comprehensive Attributes in DTN

在线阅读下载全文

作  者:崔建群[1] 余东海 常亚楠 孙佳悦 邬尧 CUI Jian-qun;YU Dong-hai;CHANG Ya-nan;SUN Jia-yue;WU Yao(School of Computer,Central China Normal University,Wuhan 430079,China)

机构地区:[1]华中师范大学计算机学院,武汉430079

出  处:《小型微型计算机系统》2022年第7期1511-1517,共7页Journal of Chinese Computer Systems

基  金:国家自然科学基金项目(61672257,61702210,61772377)资助.

摘  要:延迟容忍网络(DTN)区别于传统Internet网络,无法提供稳定端到端的连接,因此采用“存储-携带-转发”路由机制进行数据传输,而这使得消息需要长时间驻留在节点的缓存中.由于节点的缓存空间和处理能力受价格、体积和功耗的限制,因此如何对缓存进行管理成为影响路由性能的重要因素.本文根据Spray and Wait路由算法的特点,提出一种基于消息综合属性的缓存管理策略(Buffer Management Strategy Based on Message Comprehensive Attributes,MCA-BMS).该策略综合考虑消息大小、消息生存时间和消息副本数3种消息属性,确定消息的优先级,并根据优先级对消息进行转发和丢弃,同时增加了ACK确认机制,删除冗余消息,提升网络资源利用率.仿真结果表明,MCA-BMS缓存管理策略能够在消息投递率、网络开销和消息传输时延方面有明显的提升.Delay Tolerant Network(DTN),unlike traditional Internet networks,does not provide a stable end-to-end connection,so a storage-carrier-forward routing mechanism is used for data transmission,which causes messages to reside in the node's buffer for a long time.Since the buffer space and processing capacity of nodes are limited by price,volume and power consumption,how to manage the buffer becomes an important factor affecting the routing performance.According to the characteristics of the Spray and Wait routing algorithm,a Buffer Management Strategy Based on Message Comprehensive Attributes(MCA-BMS)is proposed in this paper.This strategy comprehensively considers three message attributes,namely message size,message TTL and message copy number,determines the priority of the message,forwards and discards the message according to the priority,and at the same time adds ACK-check mechanism to delete redundant messages and improve network resource utilization.The simulation results show that MCA-BMS buffer management strategy can improve the message delivery ratio,network overhead and message transmission latency.

关 键 词:DTN 消息属性 优先级 缓存管理策略 

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