Edward Hutter

  • Program Year: 4
  • Academic Institution: University of Illinois at Urbana-Champaign
  • Field of Study: Computer Science
  • Academic Advisor: Edgar Solomonik
  • Practicum(s):
    Oak Ridge National Laboratory (2019)
  • Degree(s):
    B.S. Computer Engineering, University of Illinois at Urbana-Champaign, 2017

Summary of Research

My research designs, analyzes, and implements distributed-memory algorithms and methods to better utilize machines with increasingly complex architectures. Three core design principles are emphasized: finding alternate parallel schedules for existing algorithms that minimize both intra-node and inter-node communication, exploring (conditionally) stable poly-logarithmic depth algorithms to replace those with linear critical paths (such as Gaussian elimination, Householder QR factorization, etc.), and designing robust linear and eigen-solvers on structured tensors and tensor networks with guaranteed convergence in the presence of implicit projections onto ill-favored subspaces.

I place a strong emphasis on combining both theoretical analysis of parallel costs and strong and weak scaling studies on massively parallel supercomputers in an effort to fully understand novel algorithms and parallelizations. I design tools to both track and predict critical path costs of distributed-memory algorithms.

My research focuses on improving existing algorithms and software in dense matrix factorizations, low-rank tensor factorizations, krylov solvers on tensor networks, and numerical tensor algebra.

Beyond algorithm design and implementation, my broader research interests include numerical linear algebra, parallel computing, programming models and languages, performance modeling, code generation, and software engineering. I enjoy crafting software that strikes a balance in genericity, modularity, performance, and ease of use. Specific application areas of interest lie in electronic structure calculations and quantum circuit simulation.


1. Edward Hutter and Edgar Solomonik; Accelerating Distributed-memory Autotuning via Statistical Analysis of Execution Paths, IEEE International Parallel and Distributed Processing Symposium (IPDPS), Portland, Oregon, May, 2021
2. Edward Hutter and Edgar Solomonik; Communication-avoiding Cholesky-QR2 for rectangular matrices, IEEE International Parallel and Distributed Processing Symposium (IPDPS), Rio de Jianero, Brazil, May, 2019
3. Communication-avoiding CholeskyQR2 for rectangular matrices (https://arxiv.org/abs/1710.08471)


Kenichi Miura Award (2020)
SIAM Student Travel Award 2018 (PP18)
SIAM Student Travel Award 2019 (CSE19)
SIAM Student Travel Award 2020 (PP20)