JON CALHOUN. Citizenship: United States of America Thomas M. Siebel Center for Computer Sciences 201 N Goodwin Ave Urbana, IL 61801

Transcription

1 JON CALHOUN Citizenship: United States of America Thomas M. Siebel Center for Computer Sciences 201 N Goodwin Ave Urbana, IL EDUCATION (870) jccalho2@illinois.edu University of Illinois at Urbana-Champaign August 2012 Present Ph.D. Computer Science Emphasis: High Performance and Scientific Computing, Fault Tolerance, Silent Data Corruption, Reliability, Lossy Compression Advisors: Luke Olson and Marc Snir Overall GPA: 3.73 Arkansas State University August 2008 May 2012 B.S. in Computer Science B.S. in Mathematics Minor in Statistics Summa Cum Laude and In University Honors Honors Thesis: Preemption of a CUDA Kernel Function Honors Thesis: Dynamic Contact of Viscoelastic Rods with Two Stops: Numerical Approaches Overall GPA: 4.0 WORK EXPERIENCE Argonne National Laboratory May 2016 August 2016 Research Aid Host: Franck Cappello. Explored methods for selecting error tolerances for lossy compressors. University of Illinois at Urbana-Champaign January 2015 May 2015 Teaching Assistant, CS Computer Architecture Taught multiple discussion sections and created quizzes. Lawrence Livermore National Laboratory May 2014 August 2014 Summer Intern ExMatEx Group Host: David Richards. Investigated silent data corruption detection in molecular dynamics codes. University of Illinois at Urbana-Champaign August 2012 May 2013 Teaching Assistant, CS Computer Architecture Taught multiple discussion sections, held office hours and exam review sessions, and helped develop auto-grading system. Arkansas State University May 2010 August 2012 SHADES Project Programming Intern Assisted biologists and chemists by writing GUI applications in C#, data conversion, and parsing routines. Arkansas State University January 2011 May 2011 Teaching Assistant, MATH Calculus I Graded homework for honors and regular sections.

2 RESEARCH PROJECTS Key Words: Silent Data Corruption; GPUs; Lossy Compression; Fault Injection; Error Propagation Current: As high performance computing (HPC) continues to progress, constraints on HPC system design force the handling of errors to higher levels in the software stack. Silent errors that silently corrupt system or application state are among the most troubling kinds of errors. Understanding how applications behave in the presence silent errors is critical for effective utilization of HPC systems. To this end, the LLVM based fault injection system FlipIt was created to emulate the occurrence of silent errors in HPC applications. FaultSight was developed to analysis thousands of fault injection runs preformed by FlipIt and other bit-flip style fault injectors. Combined these tools allow for efficient testing of error detection schemes. A resilient version of the linear solver algebraic multigrid (AMG) has been developed that leverages algorithmic properties of AMG in the detection and recovery schemes. Algorithmic based detection can further be enhanced by studying how error propagation at the algorithmic level as well as its implementation in code. Realizing not all errors are significant allows for optimizations that lowers power consumption the expense of adding in small errors by voltage reduction or approximation techniques such as lossy compression. Past: Previous work investigated how applications can be accelerated by porting computationally intense regions of computational biology applications and statistical analysis for industrial designs calculations to GPUs. Fault tolerance for long-running GPU computations on non-preemptable GPUs was explored with compiler based application level checkpointing for GPU kernels. In addition, the dynamic contact of viscoelastic (Kelvin-Voigt model) bodies with stationary and perfectly rigid bodies was modeled. PUBLICATIONS Exploring the Feasibility of Lossy Compression for PDE Simulations January 2017 Jon Calhoun, Luke Olson, Franck Cappello, and Marc Snir. In submission to the International Conference on Supercomputing (ICS 17). Towards a More Complete Understanding of SDC Propagation January 2017 Jon Calhoun, Marc Snir, Luke Olson, and William D. Gropp. In submission to the 26st International Symposium on High Performance Parallel & Distributed Computing (HPDC 17). IPAS: Intelligent Protection Against Silent Output Corruption in Scientific Applications March 2016 Ignacio Laguna, Martin Schulz, David F. Richards, Jon Calhoun, and Luke Olson. Published in the proceedings of the 2016 International Symposium on Code Generation and Optimization (CGO 17) in Barcelona, Spain. Towards a More Fault Resilient Multigrid Solver April 2015 Jon Calhoun, Luke Olson, Marc Snir, William D. Gropp. Published in the proceedings of the 2015 Spring Simulation Multi-Conference (Spring Sim 15) in Alexandria, VA. Fully-Implicit Ultrascale Physics Solvers and Application to Ion Source Modelling May 2014 Kris Beckwith, Seth Veitzer, Stephen F. McCormick, John W. Ruge, Luke Olson, and Jon Calhoun. Published in the IEEE Transactions on Plasma Science. FlipIt: An LLVM Based Fault Injector for HPC August 2014 Jon Calhoun, Luke Olson, and Marc Snir. Published in the proceedings of the 20th International Euro-Par Conference on Parallel Processing (Euro-Par 14) in Porto, Portugal.

3 Dynamic Contact of Viscoelastic Bodies with Two Obstacles: Mathematical and Numerical Approaches April 2013 Jeongho Ahn and Jon Calhoun. Published in the Electronic Journal of Differential Equations Vol Preemption of a CUDA Kernel Function August 2012 Jon Calhoun and Hai Jiang. Published in the proceedings of the 13th ACIS International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD 2012) in Kyoto, Japan. Acceleration of Generalized Minimum Aberration Designs of Hadamard Matrices on Graphics Processing Units June 2012 Jon Calhoun, Josh Graham, Hong Zhou, and Hai Jiang. Published in the proceedings of the Fifth International Symposium on Advances in High Performance Computing and Networking (AHPCN-2012) in Liverpool, UK. On Using a Graphics Processing Unit to Solve the Closest Substring Problem July 2011 Jon Calhoun, Josh Graham, and Hai Jiang. Published in the proceedings of the 2011 Parallel and Distributed Processing Techniques and Applications (PDPTA 11) in Las Vegas, NV. POSTERS Application of Numerical Accuracy to the Selection of Lossy Compression Error Tolerances November 2016 Presented ACM student research competition poster at Supercomputing Analyzing the Performance of a Sparse Matrix Vector Multiply for Extreme Scale Computers November 2015 Presented poster at Supercomputing Understanding the Propagation of Error Due to a Silent Data Corruption in a Sparse Matrix Vector Multiply September 2015 Presented poster at IEEE Cluster Best Poster Award Using a Graphics Processing Unit to Solve the Closest Substring Problem October 2011 Poster presentation at the Cyber Infrastructure Training Days (CI-Days) at Arkansas State. First Place CI-Days Poster Presentation. PRESENTATIONS From Detection to Optimization: Understanding Silent Error s Impact on Scientific Applications November 2016 Presented highlights of dissertation research at Supercomputing Overview of Fault Injection Techniques For HPC July 2016 Hands on lecture for 3rd Joint Laboratory for Extreme Scale Computing Summer School. Lossy Compression for HPC Checkpoint Restart: Mathematical Guidance for Error Tolerance Selection June 2016 Presented in-progress work at 5th Joint Laboratory for Extreme Scale Computing Workshop. Reducing Checkpoint Size in PlasComCM with Lossy Compression April 2016 Presented in-progress work at the 14th Annual Workshop on Charm++ and its Applications. How Silent Errors Propagate in Iterative Methods March 2016 Presented in-progress work at Fourteenth Copper Mountain Conference on Iterative Methods.

4 Evaluating Lossy Compression for HPC December 2015 Presented in-progress work at 4th Joint Laboratory for Extreme Scale Computing Workshop. Towards a More Complete Understanding of SDC Propagation July 2015 Presented in-progress research at the 3rd Joint Laboratory for Extreme Scale Computing Workshop. Effect and Propagation of Silent Data Corruption in HPC Applications May 2015 Presented Blue Waters Graduate Fellowship results at the 3rd NCSA Blue Waters Symposium. Understanding the Propagation of Error Due to a Silent Data Corruption in a Sparse Matrix Vector Multiply April, May 2015 April: Presented an invited student poster at the Salishan Conference on High-speed Computing. May: Presented poster at the 3rd NCSA Blue Waters Symposium. Understanding the Propagation of Silent Data Corruption in Algebraic Multigrid March 2015 Presented in-progress research at the Seventeenth Copper Mountain Conference on Multigrid Methods. Toward a More Fault Resilient Multigrid Solver April 2014 Presented in-progress research at the Thirteenth Copper Mountain Conference on Iterative Methods. Dynamic Contact of Viscoelastic Rods with Two Stops: Numerical Approaches March 2012 Presented at 74th Annual Meeting of OK-AR section of MAA at Henderson State University. First Place Oral Presentation Dynamic Contact of Viscoelastic Rods August 2011 Presented in-progress undergrad thesis at Mathfest 2011 in Lexington, KY Outstanding Presentation Award Closest Substring Problem July 2011 Paper presentation at PDPTA 11 in Las Vegas, NV SOFTWARE FaultSight - visualization and analysis tool for bit-flip style fault injectors. FlipIt - LLVM based fault injection system designed for HPC applications. AWARDS Outstanding Teaching Assistant - CS 233 UIUC May 2015 Blue Waters Graduate Fellowship May 2014 Outstanding Teaching Assistant - CS 398 UIUC May 2013 Arkansas Trio Program Hall of Fame February 2013 Outstanding Teaching Assistant - CS 398 UIUC January 2013 Arkansas State University Chancellor s List Fall 2008 Spring 2012 Arkansas State University Chancellor s Scholarship Fall 2008 Spring 2012 Arkansas Challenge Scholarship Fall 2008 Spring 2012 Ronald E. McNair Postbaccalaureate Achievement Program Fall 2010 Spring 2012 Outstanding Graduating Undergraduate in CS April 2012 Guido Hassin Scholarship For Science Majors Fall 2010 Spring 2011 National SMART Grant Summer 2010 SERVICE

5 UIUC Center for Innovation in Teaching & Learning Spring 2013 Spring 2017 Facilitator for Grad Academy. Trained new computer science teaching assistants. Department of Computer Science - UIUC Summer 2015 Spring 2016 Mentor for summer REU and Promoting Undergraduate Research in Engineering (PURE) students. MEMBERSHIPS Association of Computing Machinery Institute of Electrical and Electronics Engineers Mathematics Association of America Society for Industrial and Applied Mathematics