Jake Roemer, PhD

Computer Science & Engineering
Ohio State University

Dr. Jake Roemer was a graduate student, research assistant, teaching assistant, and member of the PLaSS research group at Ohio State University from 2014 to 2019. He graduated with his PhD in August 2019.

Jake's dissertation work improves data race detection in two important ways:

  • Coverage: detects more races than existing approaches
  • Efficiency: faster than existing approaches
His dissertation work resulted in three projects:
  • Raptor: an online analysis for computing Smaragdakis et al.'s causally-precedes (CP) relation
  • Vindicator: an analysis for finding more predictive races than Kini et al.'s weak-causally-precedes (WCP) while retaining soundness
  • SmartTrack: an efficient analysis for detecting predictive races substantially faster than prior predictive analyses—and nearly as fast as the state-of-the-art FastTrack happens-before analysis
Jake's contributions help make race detection effective and practical, advancing the theoretical state of the art and influencing future industrial tools and practices to make software systems more reliable.

Before grad school, Jake earned a BS in physics from Temple University in 2014. Jake was born in 1992 and grew up in Delran, New Jersey.

Jake passed away in April 2020. He was a truly creative and dedicated scholar and a wonderful friend and colleague. Jake was unusually modest and never made an academic web page, so we created this page to remember his academic accomplishments and contributions.


PhD dissertation and resulting publications

Practical High-Coverage Sound Predictive Race Detection

Jake Roemer

PhD dissertation, August 2019

SmartTrack: Efficient Predictive Race Detection

Jake Roemer, Kaan Genç, and Michael D. Bond

ACM SIGPLAN Conference on Programming Language Design and Implementation (PLDI 2020), June 2020

Extended paper     Conference paper     Source code

High-Coverage, Unbounded Sound Predictive Race Detection

Jake Roemer, Kaan Genç, and Michael D. Bond

ACM SIGPLAN Conference on Programming Language Design and Implementation (PLDI 2018), June 2018

Extended paper     Conference paper     Talk     Source code

Online Set-Based Dynamic Analysis for Sound Predictive Race Detection

Jake Roemer and Michael D. Bond

arXiv:1907.08337, July 2019

Paper     Source code

Other publications

Dependence-Aware, Unbounded Sound Predictive Race Detection

Kaan Genç, Jake Roemer, Yufan Xu, and Michael D. Bond

ACM SIGPLAN International Conference on Object-Oriented Programming, Systems, Languages, and Applications (OOPSLA 2019), October 2019

Extended paper     Conference paper     Talk     Source code

Prescient Memory: Exposing Weak Memory Model Behavior by Looking into the Future

Man Cao, Jake Roemer, Aritra Sengupta, and Michael D. Bond

ACM SIGPLAN International Symposium on Memory Management (ISMM 2016), June 2016

Paper     Talk (pptx)     Talk (pdf)     Source code

Improving Virtual Machine Migration via Deduplication

Jake Roemer, Mark Groman, Zhengyu Yang, Yufeng Wang, Chiu C. Tan, and Ningfang Mi

IEEE International Conference on Mobile Ad Hoc and Sensor Systems (MASS 2014), October 2014

Paper