ConMidbox: Consolidated Middleboxes Selection and Routing in SDN/NFV-Enabled Networks

被引：3

作者：

Liu, Guiyan ^{[1
]}

Guo, Songtao ^{[2
]}

Li, Pan ^{[1
]}

Liu, Liang ^{[1
]}

机构：

[1] Southwest Univ, Sch Elect & Informat Engn, Chongqing 400715, Peoples R China

[2] Chongqing Univ, Coll Comp Sci, Chongqing 400044, Peoples R China

来源：

2020 IEEE 34TH INTERNATIONAL PARALLEL AND DISTRIBUTED PROCESSING SYMPOSIUM IPDPS 2020 | 2020年

基金：

中国国家自然科学基金;

关键词：

Software defined networking; Consolidated middlebox; Load balancing; Rounding; Scalability; SOFTWARE-DEFINED NETWORKS; OPTIMIZATION; PLACEMENT;

D O I：

10.1109/IPDPS47924.2020.00101

中图分类号：

TP3 [计算技术、计算机技术];

学科分类号：

0812 ;

摘要：

Software defined networking (SDN) and network function virtualization (NFV) can flexibly manage software middlebox based services, and the consolidated middlebox model is able to simplify traffic routing and reduce the number of routing rules in the SDN-enabled switches. However, different network functions in middleboxes may change the volume of processed traffics, thus high congestion may occur in specific bottleneck links if middlebox selection and traffic routing are not well jointly planned. Besides, in a statically switch-controller configured SDN, traffic dynamics will not only affect the link load in data plane, but also pose a challenge to controller load balancing. Therefore, it's necessary to achieve better quality-of-service (QoS) performance in both control and data plane. This paper first formulates it as a joint traffic-aware consolidated middleboxes selection and routing (JTMSR) problem and proves its NP-hardness. Then, a two-phase RL RFRD algorithm is designed to achieve the controller and link load balancing where the first phase is to redirect selected flows by applying wildcard rules and the second phase is to find fine-grained routing path by a rounding-based algorithm with bounded approximation factor. Finally, the extensive simulation results demonstrate that the proposed algorithm has near-optimal controller load balancing and link load balancing performance and reduces response time by about 2x-5x compared with other algorithms.

引用

页码：946 / 955

页数：10

共 33 条

[1] A scalable, commodity data center network architecture [J].

Al-Fares, Mohammad ;

Loukissas, Alexander ;

Vahdat, Amin .

ACM SIGCOMM COMPUTER COMMUNICATION REVIEW, 2008, 38 (04) :63-74

[2]

[Anonymous], 2015, GUROBI OPTIMIZER REF, V6th

[3]

[Anonymous], 2012, USENIX C NETWORKED

[4]

[Anonymous], 2013, P 2 ACM SIGCOMM WORK, DOI DOI 10.1145/2491185.2491198

[5]

[Anonymous], 2015, Proceedings of the 2015 Conference on Advances

[6]

Benson T., 2010, P 10 ACM SIGC C INT, P267, DOI [DOI 10.1145/1879141.1879175, 10.1145/1879141.1879175]

[7] Deep Packet Inspection as a Service [J].

Bremler-Barr, Anat ;

Harchol, Yotam ;

Hay, David ;

Koral, Yaron .

PROCEEDINGS OF THE 2014 CONFERENCE ON EMERGING NETWORKING EXPERIMENTS AND TECHNOLOGIES (CONEXT'14), 2014, :271-282

[8]

Chang Y., 2017, 2017 Sixth Asia-Pacific Conference on AntennasandPropagation (APCAP), Xi'an, P1

[9]

Cheng GZ, 2015, 2015 IFIP NETWORKING CONFERENCE (IFIP NETWORKING)

[10]

Cohen Rami, 2015, 2015 IEEE Conference on Computer Communications (INFOCOM). Proceedings, P1346, DOI 10.1109/INFOCOM.2015.7218511

← 1 2 3 4 →