Dr. Matthew Larsen

Ph.D. Graduate
CIS Department
University of Oregon

Office:301 Deschutes
E-mail:mlarsen@cs.uoregon.edu

Bio

Matt Larsen graduated with a Ph.D. in Computer and Information Science from the University of Oregon in December of 2016. Prior to graduation, Matt accepted a full-time staff computer scientist position at Lawrence Livermore National Laboratory, and he continued in this position after graduation. Matt started the Ph.D. program in September 2013, and began his concurrent role at LLNL in January 2016. His research interests include computer graphics, GPU computing, portable performance, and scientific visualization.

Matt worked on two different research grants during his time at UO. His first project was to contribute to the NSF SI2 GraviT project, a ray tracing library for distributed-memory parallel environments. Through this work, he developed a portably-performant ray tracer that was competitive with industry standard ray tracers. This work led Matt to his second project, funded by the Department of Energy XVis grant. With this project, Matt explored the boundaries of portable performance. During his time at LLNL, Matt investigated questions on in situ rendering, which played an important role in his thesis.

Matt's research experience was enhanced by two internships: In 2014, he worked on renderers in the EAVL library at Oak Ridge Laboratory with Jeremy Meredith. Then, in 2015, he worked on lightweight in situ visualization systems at Lawrence Livermore Lab with Eric Brugger.

Honors and Awards
  • June 2016: Submission to SC16 selected as Best Paper Finalist
  • Jan. 2016: Passed UO Oral Candidacy Exam with Distinction
  • June 2015: Awarded UO's J. Donald Hubbard Scholarship
  • Dec. 2014: Passed UO Directed Research Project Exam with Distinction
  • June 2013: Graduated magna cum laude from California State University at Sacramento with a B.S. in Computer Science
First-Author Publications


Performance Modeling of In Situ Rendering
Matthew Larsen, Cyrus Harrison, James Kress, David Pugmire, Jeremy S. Meredith, and Hank Childs
The International Conference for High Performance Computing, Networking, Storage, and Analysis (SC16), Salt Lake City, UT, November 2016
Best Paper Finalist

[PDF]     [BIB]

Optimizing Multi-Image Sort-Last Parallel Rendering
Matthew Larsen, Ken Moreland, Chris Johnson, and Hank Childs
IEEE Symposium on Large Data Analysis and Visualization (LDAV), Baltimore, MD, October 2016

[PDF]     [BIB]

Strawman - A Batch In Situ Visualization and Analysis Infrastructure for Multi-Physics Simulation Codes
Matthew Larsen, Eric Brugger, Hank Childs, Jim Eliot, Kevin Griffin, and Cyrus Harrison
SC15 Workshop on In Situ Infrastructures for Enabling Extreme-scale Analysis and Visualization (ISAV-15), Austin, TX, November 2015

[PDF]     [BIB]

Volume Rendering Via Data-Parallel Primitives
Matthew Larsen, Stephanie Labasan, Paul Navratil, Jeremy Meredith, and Hank Childs
EuroGraphics Symposium on Parallel Graphics and Visualization (EGPGV), Cagliari, Italy, May 2015

[PDF]     [BIB]

Ray-Tracing Within a Data Parallel Framework
Matthew Larsen, Jeremy Meredith, Paul Navratil, and Hank Childs
IEEE Pacific Visualization, Hangzhou, China, April 2015

[PDF]     [BIB]

Other Publications


VTK-m: Accelerating the Visualization Toolkit for Massively Threaded Architectures
Ken Moreland, Chris Sewell, William Usher, Li-ta Lo, Jeremy Meredith, Dave Pugmire, James Kress, Hendrik Schroots, Kwan-Liu Ma, Matt Larsen, Hank Childs, Chun-Ming Chen, Robert Maynard, and Berk Geveci
Computer Graphics and Applications, May/June 2016

[LINK]     [BIB]

Visualization for Exascale: Portable Performance is Critical
Ken Moreland, Matt Larsen, and Hank Childs
Supercomputing Frontiers and Innovations, December 2015

[LINK]     [BIB]

Exploring Tradeoffs Between Power and Performance for a Scientific Visualization Algorithm
Stephanie Labasan, Matthew Larsen, and Hank Childs
IEEE Symposium on Large Data Analysis and Visualization (LDAV), Chicago, IL, October 2015

[PDF]     [BIB]