1 / 11

SDN Dev Group, Week 2

Wisconsin Testbed ; Design Considerations. SDN Dev Group, Week 2. Aaron Gember Aditya Akella University of Wisconsin-Madison. Development Progress. High-level design Problems encountered Lessons learned Experience with specific controllers. Wisconsin Testbed. Testbed Components.

ward
Download Presentation

SDN Dev Group, Week 2

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. Wisconsin Testbed; Design Considerations SDN Dev Group, Week 2 Aaron GemberAdityaAkella University of Wisconsin-Madison

  2. Development Progress • High-level design • Problems encountered • Lessons learned • Experience with specific controllers

  3. WisconsinTestbed

  4. Testbed Components • Switches • 3x HP ProCurve 5400 • 4x HP ProCurve 6600 • Servers • 15x Dell PowerEdge R210 (Quad-core Intel Xeon X3430, 4GB RAM, 2x 1Gbps) • 5x Dell Precision T5500 (Dual quad-core Intel Xeon E5507, 12GB RAM, 1x 1Gbps) • 5x HP dc5800 (Core 2 Duo, 2GB RAM, 2x 1Gbps)

  5. Testbed Topology CS 4th Floor East 4300E ProCurve 6600 CS 4th Floor West 4300F ProCurve 6600 CSNetwork CS 6th Floor West 6300E ProCurve 5400 CS 6th Floor West 6300F ProCurve 5400 CS 3360 (WINGS) WINGS ProCurve 5400 CS 3390 (WAIL) WAIL ProCurve 6600 CS 3390 (WAIL) WAILTWO ProCurve 6600 Internet

  6. Federation of network testbeds: OpenFlow (campuses, NLR, I2), Emulab, PlanetLab, etc. • Live resource map: http://protogeni.net/flack • GENI software stack: FlowVisor, FOAM • Experimenters create a slice + slivers at each site to reserve the resources they want

  7. Testbed for SDN Dev Group vulture lyrebird thrush eth1.2** eth1.2** eth1.2** CS 3360 (WINGS) WINGS ProCurve 5400 2 1 3 Controller wings-openflow-1

  8. Accessing the testbed • SSH to wings-openflow-1.wail.wisc.edu(from anywhere on the Internet)Username: CS username; Password: w3lcom3 • Reset your password by running passwd • Home directory is shared using NFSLocal storage on each machine at /scratch • SSH to vulture, thrush, and lyrebird to run commands on these end-hosts

  9. “Slices” for SDN Dev Group • Should reserve resources in FOAM/FlowVisor using GENI Rspecs • Statically allocated VLANs (201 – 210) • Controller should run on wings-openflow-1 and listen on port 102** • Generate traffic on vulture, lyrebird, and thrush on interface eth1.2** using IP addresses 10.2.**.0/32

  10. Design Considerations • Granularity of flow entries • Static vs. dynamic flow installation • Switch flow table size limits • Policy input • Code re-use/modular design

  11. SDN Dev Group Week 3 • Preparation • Use UW-Madison OpenFlowtestbed • Write an SDN application that relies on some supporting modules from the controller (e.g., topology, spanning tree, etc.) • Meeting topics • TBD

More Related