John C. McCullough

John and the Sunset.


Computer Science and Engineering
UC San Diego
9500 Gilman Drive, Dept 0404
La Jolla, CA 92093-0404

jmccullo ( a ) cs ucsd edu


I am a PhD Student in the Computer Science and Engineering Department at UC San Diego. I am part of the Systems and Networking Group working under Alex Snoeren.

I completed my B.S. in the Computer Science Department at Harvey Mudd College in 2006. I received my M.S. from UCSD in 2008.


scc: Cluster Storage Provisioning Informed By Application Characteristics and SLAs. Harsha Madhyastha, John C. McCullough, George Porter, Rishi Kapoor, Stefan Savage, Alex C. Snoeren, and Amin Vahdat. In Proceedings of the USENIX Conference on File and Storage Technologies, February 2012.
Evaluating the Effectiveness of Model-Based Power Characterization. John C. McCullough, Yuvraj Agarwal, Jaideep Chandrashekar, Sathyanarayan Kuppuswamy, Alex C. Snoeren, and Rajesh K. Gupta. In Proceedings of the USENIX Annual Technical Conference, 2011.
Stout: An Adaptive Interface to Scalable Cloud Storage. John C. McCullough, John Dunagan, Alec Wolman, and Alex C. Snoeren. In Proceedings of the USENIX Annual Technical Conference, June 2010.
Neon: System Support for Derived Data Management. Qing Zhang, John C. McCullough, Justin Ma, Nabil Schear, Michael Vrable, Amin Vahdat, Alex C. Snoeren, Geoff Voelker, and Stefan Savage. In Proceedings of the ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments, March 2010.
Universal Honeyfarm Containment. Jay Chen, John C. McCullough, and Alex C. Snoeren. UC San Diego, Technical Report CS2007-0902, September 2007.



The goal of scc is to model data center application requirements and the possible configurations to identify a low cost server configuration that can satisfy a given workload. Dcswitch projects page


Applying ideas from congestion control to storage to adaptively achieve a good tradeoff between latency and throughput for clients of distributed storage systems. This is work with John Dunagan and Alec Wolman at Microsoft Research.

Decongestion Control

Working with Barath Raghavan on Decongestion Control. Approaching the congested network problem with network coding that is insensitive to dropped packets.


I spent time working on the Potemkin Virtual Honeyfarm. My contributions included extending flash-cloning to fully-virtualized machines in Xen as well as re-architecting portions of the honeyfarm control code and building the new vt-enabled honeyfarm cluster. (Potemkin The Historical Object).