A Scalable and Efficient Framework for Switch Virtualization

- sponsored by National Science Foundation

Project Overview

Network virtualization is the creation of multiple coexisting logical networks, each customized to a specific purpose, by reorganizing the resources of a common physical infrastructure. Implementation of network virtualization relies on creating slices for various resources of the underlying physical infrastructure. However, there have been no effective solutions to create slices for switches, the important interconnecting components of any networks. Existing approaches to share switch bandwidth among multiple users have several drawbacks, including impractical bandwidth allocation, weak traffic isolation, poor performance guarantees, and lack of multicast support. In this project, we address the above issues, and develop a suite of bandwidth allocation and packet scheduling techniques as an integrated framework for switch virtualization, to achieve scalable, efficient, and isolated link sharing among virtual networks.

People

Deng Pan, Associate Professor

Wenrui Ma, PhD Student

Hao Jin, PhD Student (graduated in Fall 2012)

Tosmate Cheocherngngarn, PhD Student (graduated in Fall 2012)

Omair Fatmi, Master Student (graduated in Spring 2014)

Yu Li, Master Student (graduated in Summer 2013)

Jonathan Beltran, REU Student

Carlos Medina, REU Student

Dmita Levy, REU Student

Eric Jo, REU student (graduated in Summer 2014)

Arnaldo Garcia, REU Student (graduated in Spring 2013)

Alex Smith, High School Student

James Leng, High School Student

Related Publications

1. W. Ma, C. Medina, and D. Pan, ``Traffic-Aware Placement of NFV Middleboxes,'' IEEE Global Communications Conference (GLOBECOM), San Diego, CA, Dec. 2015.

2. E. Jo, L. Butler, D. Pan, and J. Liu, ``A simulation and emulation study of SDN-based multipath routing for fat-tree data center networks,'' Winter Simulation Conference (WSC), Savannah, GA, Dec. 2014.

3. O. Fatmi and D. Pan, ``Distributed multipath routing for data center networks based on stochastic traffic modeling,'' IEEE International Conference on Networking, Sensing and Control (ICNSC), Miami, FL, Apr. 2014.

4. H. Jin, D. Pan, J. Liu, and N. Pissinou, ``OpenFlow based flow level bandwidth provisioning for CICQ switches,'' IEEE Transactions on Computers, vol. 62, no. 9, pp. 1799-1812, Sep. 2013.

5. H. Jin, T. Cheocherngngarn, D. Levy, A. Smith, D. Pan, and N. Pissinou, ``Joint host-network optimization for energy-efficient data center networking,'' IEEE International Parallel and Distributed Processing Symposium (IPDPS), Boston, MA, May 2013.

6. H. Jin, D. Pan, J. Xu, and N. Pissinou, ``Efficient VM placement with multiple deterministic and stochastic resources in data centers,'' IEEE Global Communications Conference (GLOBECOM), Anaheim, CA, Dec. 2012.

7. T. Cheocherngngarn, H. Jin, J. Andrian, D. Pan, and J. Liu, ``Depth-first worst-fit search based multipath routing for data center networks,'' IEEE Global Communications Conference (GLOBECOM), Anaheim, CA, Dec. 2012.

8. D. Pan and Y. Yang, ``Flow based performance guarantee scheduling in buffered crossbar switches,'' IEEE Transactions on Communications, vol. 60, no. 12, pp. 3836-3843, Dec. 2012.

9. H. Jin, D. Pan, and N. Pissinou, ``Parallel packet switch without segmentation-and-reassembly,'' IEEE Global Communications Conference (GLOBECOM), Houston, TX, Dec. 2011.

10. T. Cheocherngngarn, J. Andrian, Z. Yang, and D. Pan, ``Queue-length proportional and max-min fair bandwidth allocation for best effort flows,'' IEEE Global Communications Conference (GLOBECOM), Houston, TX, Dec. 2011.

Related Course Developments

1. TCN-6215 Advanced Network Algorithms: Syllabus

2. TCN-5421 Theory of Networked Computation: Syllabus

3. TCN-5030 Computer Communications and Networking Technologies: Syllabus

4. CNT-4713 Net-centric Computing: Syllabus, Mininet/Floodlight practice project, programming project

Related Software