1 / 13

BalanceFlow: Controller load balancing for OpenFlow networks

BalanceFlow: Controller load balancing for OpenFlow networks. Hu, Yannan ; Wang, Wendong ; Gong, Xiangyang ; Que, Xirong ; Cheng, Shiduan Cloud Computing and Intelligent Systems (CCIS), 2012 IEEE 2nd International Conference on. Outline. Introduction Controller may be bottleneck in OpenFlow

erika
Télécharger la présentation

BalanceFlow: Controller load balancing for OpenFlow networks

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. BalanceFlow: Controller load balancing for OpenFlow networks Hu, Yannan ; Wang, Wendong ; Gong, Xiangyang ; Que, Xirong ; Cheng, Shiduan Cloud Computing and Intelligent Systems (CCIS), 2012 IEEE 2nd International Conference on

  2. Outline • Introduction • Controller may be bottleneck in OpenFlow • Existd solution • System architecture • Simulated results • Conclusion • Reference

  3. Controller may be bottleneck in OpenFlow • If a OpenFlow switch get a packet which does not match the flow table, OpenFlow switch will send the packet to controller.

  4. Controller may be bottleneck in OpenFlow

  5. Existd mechanism- OpenFlow 1.2 • OpenFlow 1.2 supports multiple controller connection for one switch. • Each controller may send a OFPT_ROLE_REQUEST message to communicate its role to the switch • OFPCR_ROLE_MASTER • OFPCR_ROLE_SLAVE • But there is not any mechanism for the communication and flow-request load balancing.

  6. System architecture • BalanceFlow

  7. Controller’s job • Each controller k maintains an N × N matrix M k, where N is the number of switches in the network. The element in the ith row, jth column denotes the average number of flow-requests from switch i to switch j . • The super controller is responsible for collecting the flow-requests matrixes published by all controllers.

  8. Detecting imbalance • The super controller detects controller load imbalance when the average number of flow-requests handled by any controller contributes more than some threshold of the total flow-requests rate in the network.

  9. Generating allocation information • Algorigthm • Cost :

  10. Simulated results

  11. Conclusion • The super controller runs partition algorithm and reallocates the load of different controller by distributing allocation rules to switches. • BalanceFlow can handle controller load balancing entirely at the controller level.

  12. Reference • OpenFlow Switch Specification 1.3.2 - Open Networking Foundation

More Related