Nathan Otterness

About

I obtained my Ph.D. in August of 2022 from the Computer Science department at the University of North Carolina at Chapel Hill. I worked under Professor James Anderson in the Real-time Systems group. My research primarily focused on enabling predictable timing in programs that concurrently share GPUs. I have also contributed to projects involving shared-hardware isolation (on CPUs) and mixed-criticality scheduling.

Apart from my research, my personal interests include operating systems, a wide variety of low-level programming topics, and fractals (High-resolution examples include this and this. Here's a cropped and downscaled version of one.). I spend a large amount of time programming as a hobby. My github page includes most of my more-complete programming projects. I was a winner of the 2020 International Obfuscated C Code Contest.

As of September 2022 I work at NVIDIA, in the hardware infrastructure group on projects related to GPU and datacenter simulation and testing. I previously worked at NVIDIA for three summer internships: May through July of 2018, May through August of 2019, and May through August of 2021. From 2012 through 2015 I worked in computer security research under Professor Fabian Monrose. As an undergraduate, I spent a few months as a Co-Op at IBM.

Teaching

Spring 2019: I taught COMP 455: Models of Languages and Computation (aka "Automata Theory"). Link to the course website.

Publications

• Tyler Yandrofski, Jingyuan Chen, Nathan Otterness, James Anderson, and F. D. Smith. Making Powerful Enemies on NVIDIA GPUs, Real-Time Systems Symposium (RTSS), December 2022. link
• Nathan Otterness. Developing Real-Time GPU-Sharing Platforms for Artificial-Intelligence Applications, Dissertation, University of North Carolina at Chapel Hill, August 2022. link
• Nathan Otterness, James Anderson. Exploring AMD GPU Scheduling Details by Experimenting With "Worst Practices", Real-Time Systems, March 2022. (This is an extended version of the RTNS 2021 paper for journal publication.) link
• Nathan Otterness, James Anderson. Exploring AMD GPU Scheduling Details by Experimenting With "Worst Practices", International Conference on Real-Time Networks and Systems (RTNS), April 2021. Best paper award. link
• Nathan Otterness, James Anderson. AMD GPUs as an Alternative to NVIDIA for Supporting Real-Time Workloads, Euromicro Conference on Real-Time Systems (ECRTS), July 2020. link
• Ming Yang, Tanya Amert, Kecheng Yang, Nathan Otterness, James Anderson, F. D. Smith, and Shige Wang. Making OpenVX Really "Real-Time", IEEE Real-Time Systems Symposium (RTSS), December 2018. link
• Namhoon Kim, Stephen Tang, Nathan Otterness, James H. Anderson, F. Donelson Smith, and Donald E. Porter. Supporting I/O and IPC via Fine-Grained OS Isolation for Mixed-Criticality Real-Time Tasks, Conference on Real-Time Networks and Systems (RTNS), October 2018. Outstanding paper award. link
• Ming Yang, Nathan Otterness, Tanya Amert, Joshua Bakita, James Anderson, and F. D. Smith. Avoiding Pitfalls when Using NVIDIA GPUs for Real-Time Tasks in Autonomous Systems, Euromicro Conference on Real-Time Systems (ECRTS), July 2018. link
• Joshua Bakita, Nathan Otterness, James Anderson, and F. D. Smith. Scaling Up: The Validation of Empirically Derived Scheduling Rules on NVIDIA GPUs, Workshop on Operating Systems Platforms for Embedded Real-Time Applications (OSPERT), July 2018. link
• Tanya Amert, Nathan Otterness, Ming Yang, James Anderson, and F. D. Smith. GPU Scheduling on the NVIDIA TX2: Hidden Details Revealed, IEEE Real-Time Systems Symposium (RTSS), December 2017. link
• Micaiah Chisholm, Namhoon Kim, Stephen Tang, Nathan Otterness, James Anderson, F. D. Smith, and Don Porter. Supporting Mode Changes while Providing Hardware Isolation in Mixed-Criticality Multicore Systems, Conference on Real-Time Networks and Systems (RTNS), October 2017. link
• Nathan Otterness, Ming Yang, Tanya Amert, James Anderson, and F. D. Smith. Inferring the Scheduling Policies of an Embedded CUDA GPU, Workshop on Operating Systems Platforms for Embedded Real-Time Applications (OSPERT), June 2017. link
• Nathan Otterness, Ming Yang, Sarah Rust, Eunbyung Park, James Anderson, F. D. Smith, Alex Berg, and Shige Wang. An Evaluation of the NVIDIA TX1 for Supporting Real-Time Computer-Vision Workloads, Real-Time and Embedded Technology and Applications Symposium (RTAS), April 2017. link
• Namhoon Kim, Micaiah Chisholm, Nathan Otterness, James Anderson, and F. D. Smith. Allowing Shared Libraries while Supporting Hardware Isolation in Multicore Real-Time Systems, Real-Time and Embedded Technology and Applications Symposium (RTAS), April 2017. link
• Micaiah Chisholm, Namhoon Kim, Bran Ward, Nathan Otterness, James Anderson, and F. D. Smith. Reconciling the Tension Between Hardware Isolation and Data Sharing in Mixed-Criticality, Multicore Systems, IEEE Real-Time Systems Symposium (RTSS), December 2016. Best student paper award. link
• Nathan Otterness, Vance Miller, Ming Yang, James Anderson, and F. D. Smith. GPU Sharing for Image Processing in Embedded Real-Time Systems, Workshop on Operating Systems Platforms for Embedded Real-Time Applications (OSPERT), July 2016. link
• Jan Werner, George Baltas, Rob Dallara, Nathan Otterness, Kevin Z. Snow, Fabian Monrose, and Michalis Polychronakis. No-Execute-After-Read: Preventing Code Disclosure in Commodity Software, Asia Conference on Computer and Communications Security (AsiaCCS), June 2016. link
• Teryl Taylor, Kevin Z. Snow, Nathan Otterness, and Fabian Monrose. Cache, Trigger, Impersonate: Enabling Context-Sensitive Honeyclient Analysis On-the-Wire, ISOC Network and Distributed Systems Security Symposium (NDSS), February 2016. link
• Blaine Stancill, Kevin Z. Snow, Nathan Otterness, Fabian Monrose, Lucas Davi, and Ahmad-Reza Sadeghi. Check My Profile: Leveraging Static Analysis for Fast and Accurate Detection of ROP Gadgets, Research in Attacks, Intrusions and Defenses (RAID) Symposium, October 2013. link