I am the tech lead of the CORD platform team, the foundation that all the CORD use cases (R-CORD, M-CORD, E-CORD, A-CORD) are building on.
Prior to joining the CORD project in 2015, I built academic research infrastructure for the distributed systems and networking communities, specifically the PlanetLab, VINI, VICCI, and GENI testbeds. Many of these projects were influential in the development of SDN; for an overview, see the ACM Queue article, “The Road to SDN: An intellectual history of programmable networks”. I served on the GENI Architecture Board and am currently a member of the Chameleon Cloud Science Advisory Board.
How long have been working in the CORD community?
I have been working on CORD since the start of the project in 2015.
What contributions have you made in the past to the CORD community?
A key challenge of CORD is to balance the requirements of operators vs. contributors. Operators need a production-ready CORD reference implementation running on real hardware; contributors want environments that are useful for developing, testing, and debugging CORD. My contributions to CORD revolve around trying to advance both of these agendas in parallel. At the beginning of the project, I took the lead on integrating the various components of CORD (ONOS, XOS, OpenStack, MaaS) into a working system. I am the primary owner of the current CORD build system. I also develop CORD-in-a-Box, a useful environment for integration testing, system development, and getting initial hands-on experience with CORD using a single server. I’ve demonstrated the progress we’ve been making with CORD at the last three ONS conferences.
What are you actively working on in CORD?
I am currently working on refactoring the CORD build system to make life easier for developers who want to build CORD services or contribute to the platform. I’m also the mentor of the Performance Brigade, which is integrating DPDK support into CORD.
Why do you feel you would be a good candidate for this position?
I am a practical system builder rather than a soaring visionary.
I have a pretty deep understanding of the pieces of CORD and how they fit together.
I can see both the exciting potentials and current warts of the CORD platform.
I care a great deal about making CORD easier to use, understand, and contribute to.
My prior work in academic testbeds such as PlanetLab has given me experience building real systems and fostering communities.
Are there any changes you would like to bring to the community if elected into this position?
I believe that the CORD project provides fertile ground for collaboration between industry and academia. Academics need real problems to work on, as well as exposure to the methodologies that go into building production systems; industry can benefit from fresh perspectives and longer-term thinking; the CORD project has all of these. If re-elected I will try to draw more academics and students into the CORD community. I also will look for opportunities for CORD to leverage research infrastructure like CloudLab (already big supporters of CORD) and influence new research testbeds.