• Skip to main content
  • Skip to after header navigation
  • Skip to site footer
ERN: Emerging Researchers National Conference in STEM

ERN: Emerging Researchers National Conference in STEM

  • About
    • About AAAS
    • About the NSF
    • About the Conference
    • Partners/Supporters
    • Project Team
  • Conference
  • Abstracts
    • Undergraduate Abstract Locator
    • Graduate Abstract Locator
    • Abstract Submission Process
    • Presentation Schedules
    • Abstract Submission Guidelines
    • Presentation Guidelines
  • Travel Awards
  • Resources
    • Award Winners
    • Code of Conduct-AAAS Meetings
    • Code of Conduct-ERN Conference
    • Conference Agenda
    • Conference Materials
    • Conference Program Books
    • ERN Photo Galleries
    • Events | Opportunities
    • Exhibitor Info
    • HBCU-UP/CREST PI/PD Meeting
    • In the News
    • NSF Harassment Policy
    • Plenary Session Videos
    • Professional Development
    • Science Careers Handbook
    • Additional Resources
    • Archives
  • Engage
    • Webinars
    • ERN 10-Year Anniversary Videos
    • Plenary Session Videos
  • Contact Us
  • Login

Policy-Aware, Energy-Efficient Middlebox Placement in SDN-Enabled Data Centers

Graduate #54
Discipline: Computer Sciences and Information Management
Subcategory: Computer Science & Information Systems

Alexander Ing - California State University Dominguez Hills
Co-Author(s): Daniel Hernandez, Bin Tang, and Mohsen Beheshti, California State University Dominguez Hills, Carson, CA



A middlebox, also called network appliance, is a computer networking device that transforms, inspects, filters, or otherwise manipulates traffic for purposes other than packet forwarding. Examples of middlebox includes firewalls, load balancers, intrusion detection/prevention systems. Traditional middlebox hardware is widely deployed in enterprise networks to improve network security and performance. However, deploying middleboxes is currently a hard and complex problem, which involves network operator to manually install them inside networks. This is neither flexible nor efficient, and error-prone. Data centers consist of tens of thousands of server machines that support a large number of Internet services such as social networking, video streaming, and search engines. Data center policies require that VM traffic in data centers traverse a sequence of specified middleboxes, for the purposes of security and performance. In cloud data center, the integration of Software Defined Network (SDN) and Network Function Virtualization has been recently proposed to enables efficient placement of software-based middblebox in commercial off the-shelf switches. There are a few studies that address such middlebox placement problem, however, none of them are from an algorithmic angle. In this work, we first formulate this problem formally. We show that this problem is NP-hard. We then propose two time-efficient heuristic algorithms, one is Random, the other is Greedy. We show via extensive simulations that Greedy always outperforms Random in all different network scenarios. As a future work, we would like to take into consideration of load balancing. That is, the VM communication traffic on different switches and middleboxes should be equally distributed, without overloading any of the particular device.

Funder Acknowledgement(s): This research is funded in part through the National Science Foundation (NSF) under grant #HRD-1302873 and the Chancellor's Office of the California State University. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation or the Chancellor's Office of the CSU.

Faculty Advisor: Bin Tang, btang@csudh.edu

Role: My contribution to this research was the creation of the K-ary fat tree simulation, allowing us to change parameters within the simulated datacenter; this robust simulation was used to test the efficiency of the middlebox placement algorithms. Also, I participated in the collection of data through the experimentation on both the random and greedy algorithms. Once the data was collected I also helped analyze and interpret the results of the data from the experiments.

Sidebar

Abstract Locators

  • Undergraduate Abstract Locator
  • Graduate Abstract Locator

This material is based upon work supported by the National Science Foundation (NSF) under Grant No. DUE-1930047. Any opinions, findings, interpretations, conclusions or recommendations expressed in this material are those of its authors and do not represent the views of the AAAS Board of Directors, the Council of AAAS, AAAS’ membership or the National Science Foundation.

AAAS

1200 New York Ave, NW
Washington,DC 20005
202-326-6400
Contact Us
About Us

  • LinkedIn
  • Facebook
  • Instagram
  • Twitter
  • YouTube

The World’s Largest General Scientific Society

Useful Links

  • Membership
  • Careers at AAAS
  • Privacy Policy
  • Terms of Use

Focus Areas

  • Science Education
  • Science Diplomacy
  • Public Engagement
  • Careers in STEM

Focus Areas

  • Shaping Science Policy
  • Advocacy for Evidence
  • R&D Budget Analysis
  • Human Rights, Ethics & Law

© 2023 American Association for the Advancement of Science