About Me

I am a Research Computer Scientist at the USC Information Sciences Institute in Arlington, VA, USA. My research interests include self-aware adaptive systems, self-adaptive/goal-oriented software, and fault-tolerant/long-lived systems. I am particularly interested in scheduling problems related to performance, power, and accuracy tradeoffs in systems ranging in scale from embedded/mobile/IoT to HPC/cloud.

I received my Ph.D. in Computer Science from the University of Chicago under the supervision of Hank Hoffmann. You can find a summary of my graduate research here. Prior to graduate school, I was a software engineer at Lockheed Martin for five years.

Recent News

2024

Highlights

  • Our paper "Timely Wildfire Perimeter Mapping for Unmanned Aerial Platforms" was presented and published in the 2024 Optica Imaging Congress (due to a scheduling conflict in 2023).
  • Our paper "Evaluating Deep Learning Recommendation Model Training Scalability with the Dynamic Opera Network" was accepted to The 4th Workshop on Machine Learning and Systems (EuroMLSys).
  • I've been promoted to Research Computer Scientist!

Software Releases

2023

Highlights

  • We shared the Best Paper Award at FMEC 2023!
  • Our extended abstract "Deep Learning Recommendation Model Training Co-design with the Dynamic Opera Network" was accepted to the IEEE Conference on High Performance Extreme Computing (HPEC 2023) for a poster presentation.
  • Our paper "Distributed Edge Machine Learning Pipeline Scheduling with Reverse Auctions" (a collaboration with Dave King at AFIT) was accepted to the IEEE International Conference on Fog and Mobile Edge Computing (FMEC 2023).
  • Our summary paper "Timely Wildfire Perimeter Mapping for Unmanned Aerial Platforms" was accepted for presentation at the 2023 Optica Imaging Congress.
  • Our "QuantPipe" paper on using post-training quantization to adapt to dynamic network conditions in distributed transformer pipelines was accepted to the 2023 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP).

Software Releases

2022

Highlights

  • Pervaiz et. al.'s paper "GOAL: Supporting General and Dynamic Adaptation in Computing Systems" was accepted to The ACM SIGPLAN Symposium on New Ideas, New Paradigms, and Reflections on Programming and Software (SPLASH Onward! 2022).
  • Our "PipeEdge" paper on pipelining large machine learning transformers across heterogeneous edge devices was accepted to Euromicro Digital System Design (DSD) 2022 special session on Applications, Architectures, Methods and Tools for Machine – and Deep Learning (AAMTM).

Software Releases

2021

Highlights

  • Our CASPER project paper on using the Halide programming language for HPC Synthetic Aperture Radar processing received an Outstanding Paper Award at HPEC 2021!
  • Oak Ridge National Lab published an article about our CASPER project. The article was also picked up by HPCwire.

Software Releases

2020

Highlights

  • Our paper on characterizing a large-scale NUMA shared memory system (an HPE SuperDome Flex) was accepted to the HiPar workshop at SC20 (Supercomputing 2020).
  • Our paper on CASPER's design was accepted to the ESPM2 workshop at SC20 (Supercomputing 2020).

Software Releases

2019

Highlights

  • We are working on CASPER (Compiler Abstractions Supporting high Performance on Extreme-scale Resources) for the DARPA PAPPA microexploration program.
  • Our "CoPPer" paper received the Karsten Schwan Best Paper Award at ICAC 2019!
  • Our paper Proteus: Language and Runtime Support for Self-Adaptive Software Development appeared in the March-April edition of IEEE Software.

Software Releases

2018

Highlights

  • As part of the NASA High Performance Spaceflight Computing project, we are designing and prototyping the systems sofware for a heterogeneous multicore fault-tolerant System-on-Chip.
  • I am presenting our paper Energy-efficient Application Resource Scheduling using Machine Learning Classifiers at ICPP 2018.
  • I am now working at USC/ISI in Arlington, VA!

Software Releases

Publications

2024

Andrew Rittenbach, Connor Imes, and John Paul Walters. Timely Wildfire Perimeter Mapping for Unmanned Aerial Platforms. In: Optica Imaging Congress 2024 (3D, AOMS, COSI, ISA, pcAOP). 2024.

Connor Imes, Andrew Rittenbach, Peng Xie, Dong In D. Kang, John Paul Walters, Stephen P. Crago. Evaluating Deep Learning Recommendation Model Training Scalability with the Dynamic Opera Network. In: EuroMLSys. 2024.

2023

Connor Imes, David W. King, John Paul Walters. Distributed Edge Machine Learning Pipeline Scheduling with Reverse Auctions. In: FMEC. 2023. [Best Paper Award]

Haonan Wang, Connor Imes, Souvik Kundu, Peter A. Beerel, Stephen P. Crago, John Paul Walters. QuantPipe: Applying Adaptive Post-Training Quantization for Distributed Transformer Pipelines in Dynamic Edge Environments. In: ICASSP. 2023.

2022

Ahsan Pervaiz, Yao Hsiang Yang, Adam Duracz, Ferenc Bartha, Ryuichi Sai, Connor Imes, Robert Cartwright, Krishna Palem, Shan Lu, and Henry Hoffmann. GOAL: Supporting General and Dynamic Adaptation in Computing Systems. In: Onward! 2022.

Yang Hu, Connor Imes, Xuanang Zhao, Souvik Kundu, Peter A. Beerel, Stephen P. Crago, John Paul Walters. PipeEdge: Pipeline Parallelism for Large-Scale Model Inference on Heterogeneous Edge Devices. In: DSD. 2022.

2021

Connor Imes, Tzu-Mao Li, Mark Glines, Rishi Khan, John Paul Walters. Distributed and Heterogeneous SAR Backprojection with Halide. In: HPEC. 2021. [Outstanding Paper Award]

2020

Connor Imes, Alexei Colin, Naifeng Zhang, Ajitesh Srivastava, Viktor Prasanna, John Paul Walters. Compiler Abstractions and Runtime for Extreme-scale SAR and CFD Workloads. In: ESPM2. 2020.

Connor Imes, Steven Hofmeyr, Dong In D. Kang, John Paul Walters. A Case Study and Characterization of a Many-socket, Multi-tier NUMA HPC Platform. In: HiPar. 2020.

2019

Connor Imes, Huazhe Zhang, Kevin Zhao, and Henry Hoffmann. CoPPer: Soft Real-time Application Performance Using Hardware Power Capping. In: ICAC. 2019. [Best Paper Award]

Saeid Barati, Ferenc A. Bartha, Swarnendu Biswas, Robert Cartwright, Adam Duracz, Donald Fussell, Henry Hoffmann, Connor Imes, Jason Miller, Nikita Mishra, Arvind, Dung Nguyen, Krishna V. Palem, Yan Pei, Keshav Pingali, Ryuichi Sai, Andrew Wright, Yao-Hsiang Yang, Sizhuo Zhang. Proteus: Language and Runtime Support for Self-Adaptive Software Development. In: IEEE Software, vol. 36, no. 2, pp. 73-82. March-April 2019.

2018

Connor Imes, Steven Hofmeyr, and Henry Hoffmann. Energy-efficient Application Resource Scheduling using Machine Learning Classifiers. In: ICPP. 2018.

Nikita Mishra, Connor Imes, John Lafferty, and Henry Hoffmann. Controlling AI Engines in Dynamic Environments. In: SysML Conference. 2018.

Nikita Mishra, Connor Imes, John Lafferty, and Henry Hoffmann. CALOREE: Learning Control for Predictable Latency and Low Energy. In: ASPLOS. 2018.

2016

Connor Imes, Lars Bergstrom, and Henry Hoffmann. A Portable Interface for Runtime Energy Monitoring. In: FSE. 2016.

Connor Imes, David H. K. Kim, Martina Maggio, and Henry Hoffmann. Portable Multicore Resource Management for Applications with Performance Constraints. In: MCSoC. 2016.

Connor Imes and Henry Hoffmann. Bard: A Unified Framework for Managing Soft Timing and Power Constraints. In: SAMOS. 2016.

2015

David H. K. Kim, Connor Imes, and Henry Hoffmann. Racing and Pacing to Idle: Multicore Energy Optimization Under Performance Constraints. In: CPSNA. 2015.

Connor Imes, David H. K. Kim, Martina Maggio, and Henry Hoffmann. POET: A Portable Approach to Minimizing Energy Under Soft Real-time Constraints. In: RTAS. 2015.

2014

Connor Imes and Henry Hoffmann. Minimizing Energy Under Performance Constraints on Embedded Platforms: Resource Allocation Heuristics for Homogeneous and Single-ISA Heterogeneous Multi-Cores. In: SIGBED Review. 2014.

Education

Ph.D., Computer Science, University of Chicago, 2018.
Dissertation: Balancing Performance and Energy in Computing Systems [pdf]
Committee: Henry Hoffmann, Shan Lu, Steven Hofmeyr

M.S., Computer Science, University of Chicago, 2015.
Thesis: Managing Diversity in Performance and Energy Characteristics on Embedded Systems [pdf]

B.S., Computer Engineering and Computer Science, University of Southern California, 2008.

Awards & Recognitions

Research

  • Best Paper Award - The Eighth IEEE International Conference on Fog and Mobile Edge Computing (FMEC), 2023.
  • Outstanding Paper Award - IEEE Conference on High Performance Extreme Computing (HPEC), 2021.
  • Karsten Schwan Best Paper Award - The 16th IEEE International Conference on Autonomic Computing (ICAC), 2019.
  • IEEE Micro Top Pick Honorable Mention - For our ASPLOS 2018 "CALOREE" paper, 2019.
  • Best Paper Award - The 4th Embedded Operating Systems Workshop (EWiLi), 2014.
  • Honorable Mention Award - Federal Aviation Administration, Undergraduate Senior Design Project: Runway Safety / Runway Incursions Challenge FAA Design Competition for Universities, 2008.

Industry

  • Certificate of Appreciation - MAJ Hans Baltzer, NLD Army, Chairman - Programme Management Group, Multilateral Interoperability Programme, 2013.
  • Certificate of Appreciation - LTC Brian Lyttle, U.S. Army, Product Manager, Strategic Mission Command, 2012.
  • SPOT Award - Lockheed Martin, 2010.
  • Certificate of Achievement - Alan Chester, Engineering Site Manager, Lockheed Martin, 2010.
  • Certificate of Appreciation - Venis Knight, Product Manager, Tactical Battle Command, 2010.