一种高效的OpenFlow流表拆分压缩算法
|
摘要
在软件定义网络中,OpenFlow通过丰富的匹配字段提供了网络流的细粒度管理,同时也带来了流表规模庞大的问题,给OpenFlow交换机的TCAM存储资源提出了严峻的挑战.为此,本文基于匹配域间关系,提出一种高效的OpenFlow流表拆分压缩算法.该算法首先分析了流表匹配字段之间的共存和互斥关系,并据此将流表划分出多个规模较小的子流表,进而针对每个字段建立判定条件,对子流表做进一步压缩,以实现OpenFlow流表的高效存储.最后,借助实际网络流量样本,对本文所提算法的流表压缩性能进行了实验评估.实验结果表明:本文所提算法的流表压缩率明显高于现有压缩算法,可有效节省流表存储资源.
In software-defined networking,OpenFlow offers fine-grained management to network flows with abundant matching fields,which results in explosive growth of flow table size and imposes great challenges on TCAM storage resource in OpenFlow switches.To solve this problem,this paper presents an efficient OpenFlow flow table splitting and compressing algorithm by exploiting the relationships among matching fields. Firstly,the paper analyzes coexistence and conflict relationships among matching fields,and divides a given flow table into multiple sub-flow tables. Then,we compress sub-flow tables by building a condition for each field. Finally,we evaluate the flow table compression performance of our proposed algorithm in virtue of real network traffic traces by experiments. The experimental results indicate that our proposed algorithm performs better than existing compression algorithms in terms of flow table storage compression rate,and effectively saves TCAM storage resources.
In software-defined networking,OpenFlow offers fine-grained management to network flows with abundant matching fields,which results in explosive growth of flow table size and imposes great challenges on TCAM storage resource in OpenFlow switches.To solve this problem,this paper presents an efficient OpenFlow flow table splitting and compressing algorithm by exploiting the relationships among matching fields. Firstly,the paper analyzes coexistence and conflict relationships among matching fields,and divides a given flow table into multiple sub-flow tables. Then,we compress sub-flow tables by building a condition for each field. Finally,we evaluate the flow table compression performance of our proposed algorithm in virtue of real network traffic traces by experiments. The experimental results indicate that our proposed algorithm performs better than existing compression algorithms in terms of flow table storage compression rate,and effectively saves TCAM storage resources.
引文
[1]Kannan K,Banerjee S.Compact TCAM:flow entry compaction in TCAM for pow er aw are SDN[C].In:Proceedings of the Distributed Computing and Netw orking,Springer Berlin Heidelberg,2013:439-444.
[2]Banerjee S,Kannan K.Tag-In-Tag:efficient flow table management in SDN sw itches[C].In:Proceedings of IEEE International Conference on Netw ork and Service M anagement(CNSM),Rio de Janeiro,Brazil,2014:109-117.
[3]Bedhiaf I L,Burguin R,Cherkaoui O,et al.TCAM table resource allocation for virtual openflow sw itch[C].In:Proceedings of the Annual M editerranean on Ad Hoc Netw orking Workshop(M edHoc),Ajaccio,2013:128-132.
[4]Veeramani S,Kumar M,Mahammad S N.Minimization of flow table for TCAM based openflow sw itches by virtual compression approach[C].In Proceedings of Advanced Netw orks and Telecommuncations Systems(ANTS),Kattankulathur,2013:1-4.
[5]Guo Z,Xu Y,Cello M,et al.Jump Flow:reducing flow table usage in softw are-defined netw orks[J].Computer Netw orks,2015,92(2):300-315.
[6]Leng B,Huang L,Wang X,et al.A mechanism for reducing flow tables in softw are defined netw ork[C].In:Proceedings of the 2015IEEE International Conference on Communications(ICC),London,2015:5302-5307.
[7]Luo S,Yu H,Li L M.Fast incremental flow table aggregation in SDN[C].In:Proceedings of the International Conference on Computer Communication and Netw orks,Shanghai,2014:1-8.
[8]Luo S,Yu H,Li L.Practical flow table aggregation in SDN[J].Computer Netw orks the International Journal of Computer&Telecommunications Netw orking,2015,92(P1):72-88.
[9]Ge J,Chen Z,Wu Y,et al.H-SOFT:a heuristic storage space optimisation algorithm for flow table of Open Flow[J].Concurrency&Computation Practice&Experience,2015,27(13):3497-3509.
[10]E Yue-peng,Chen Zhi,Ge Jing-guo.An efficient implementation of storage and lookup for flow tables in Open Flow[J].Science China:Information Science,2015,45(10):1280-1288.
[11]Liu Zhong-jin,Li Yong,Su Li,et al.TCAM-efficient flow table mapping scheme for Open Flow multiple-table pipelines[J].Tsinghua Univ(Sci&Technol),2014,54:437-442.
[12]Zuo Qing-yun,Chen Ming,Zhao Guang-song,et al.Research on Open Flow-based SDN technologies[J].Journal of Softw are,2013,24(5):1078-1097.
[10]鄂跃鹏,陈智,葛敬国,等.一种高效的Open Flow流表存储与查找实现方法[J].中国科学:信息科学,2015,45(10):1280-1288.
[11]刘中金,李勇,苏厉,等.TCAM存储高效的Open Flow多级流表映射机制[J].清华大学学报(自然科学版),2014,54(4):437-442.
[12]左青云,陈鸣,赵广松,等.基于Open Flow的SDN技术研究[J].软件学报,2013,24(5):1078-109.
[2]Banerjee S,Kannan K.Tag-In-Tag:efficient flow table management in SDN sw itches[C].In:Proceedings of IEEE International Conference on Netw ork and Service M anagement(CNSM),Rio de Janeiro,Brazil,2014:109-117.
[3]Bedhiaf I L,Burguin R,Cherkaoui O,et al.TCAM table resource allocation for virtual openflow sw itch[C].In:Proceedings of the Annual M editerranean on Ad Hoc Netw orking Workshop(M edHoc),Ajaccio,2013:128-132.
[4]Veeramani S,Kumar M,Mahammad S N.Minimization of flow table for TCAM based openflow sw itches by virtual compression approach[C].In Proceedings of Advanced Netw orks and Telecommuncations Systems(ANTS),Kattankulathur,2013:1-4.
[5]Guo Z,Xu Y,Cello M,et al.Jump Flow:reducing flow table usage in softw are-defined netw orks[J].Computer Netw orks,2015,92(2):300-315.
[6]Leng B,Huang L,Wang X,et al.A mechanism for reducing flow tables in softw are defined netw ork[C].In:Proceedings of the 2015IEEE International Conference on Communications(ICC),London,2015:5302-5307.
[7]Luo S,Yu H,Li L M.Fast incremental flow table aggregation in SDN[C].In:Proceedings of the International Conference on Computer Communication and Netw orks,Shanghai,2014:1-8.
[8]Luo S,Yu H,Li L.Practical flow table aggregation in SDN[J].Computer Netw orks the International Journal of Computer&Telecommunications Netw orking,2015,92(P1):72-88.
[9]Ge J,Chen Z,Wu Y,et al.H-SOFT:a heuristic storage space optimisation algorithm for flow table of Open Flow[J].Concurrency&Computation Practice&Experience,2015,27(13):3497-3509.
[10]E Yue-peng,Chen Zhi,Ge Jing-guo.An efficient implementation of storage and lookup for flow tables in Open Flow[J].Science China:Information Science,2015,45(10):1280-1288.
[11]Liu Zhong-jin,Li Yong,Su Li,et al.TCAM-efficient flow table mapping scheme for Open Flow multiple-table pipelines[J].Tsinghua Univ(Sci&Technol),2014,54:437-442.
[12]Zuo Qing-yun,Chen Ming,Zhao Guang-song,et al.Research on Open Flow-based SDN technologies[J].Journal of Softw are,2013,24(5):1078-1097.
[10]鄂跃鹏,陈智,葛敬国,等.一种高效的Open Flow流表存储与查找实现方法[J].中国科学:信息科学,2015,45(10):1280-1288.
[11]刘中金,李勇,苏厉,等.TCAM存储高效的Open Flow多级流表映射机制[J].清华大学学报(自然科学版),2014,54(4):437-442.
[12]左青云,陈鸣,赵广松,等.基于Open Flow的SDN技术研究[J].软件学报,2013,24(5):1078-109.