Kazem Cheshmi

I am a PhD candidate at ParaMathics lab, at the department of computer science, University of Toronto, and advised by Prof. Maryam Mehri Dehnavi. My research area includes sparse computations, high-performance computing, compiler optimization, and parallel programming.
I am in the job market.
I am teaching CSC367 parallel computing in Winter 2020. Drop me an email if you are on the waitlist.

Education

  • Ph.D., Computer Scince - University of Toronto

  • M.A.Sc., Computer Engineering - University of Tehran

  • B.Eng., Computer Engineering - Ferdowsi University of Mashhad

Major Awards

  • Recipient of ACM-IEEE CS George Michael Memorial HPC fellowship. (2020)
  • First place in ACM student research competition grand finals across all ACM conferences. (2017)
  • Recipient of Adobe fellowship. (2018)
  • Recipient of Wolfond fellowship, computer science, University of Toronto. (2018)
  • Best poster paper award in HPCS conference. (2018)

  • First place in ACM CGO student research competition. (2017)
  • German Academic Exchange Service (DAAD) to attend the University of Rostock, Germany. (2011)

Publications

  • Kazem Cheshmi, Danny M. Kaufman, Shoaib Kamil, and Maryam Mehri Dehnavi. "NASOQ: Numerically Accurate Sparsity-Oriented QP Solver." ACM Transactions on Graphics , SIGGRAPH'20 .
  • Bangtian Liu*, Kazem Cheshmi*, Saeed Soori, Michelle Mills Strout, Maryam Mehri Dehnavi. "MatRox: Modular approach for improving data locality in Hierarchical (Mat)rix App(Rox) imation". PPoPP'20. (*Equal contributions)
  • Mahdi Soltan Mohammadi, Tomofumi Yuki, Kazem Cheshmi, Eddie C Davis, Mary Hall, Maryam Mehri Dehnavi, Payal Nandy, Catherine Olschanowsky, Anand Venkat, Michelle Mills Strout. "Sparse computation data dependence simplification for efficient compiler-generated inspectors." PLDI'19.
  • Kazem Cheshmi, S. Kamil, Michelle Mills Strout, and Maryam Mehri Dehnavi. "ParSy: inspection and transformation of sparse matrix computations for parallelism." SC'18.
  • Kazem Cheshmi, Shoaib Kamil, Michelle Mills Strout, and Maryam Mehri Dehnavi. "Sympiler: Transforming Sparse Matrix Codes by Decoupling Symbolic Analysis." SC'17 .
  • Mahdi Soltan Mohammadi, Kazem Cheshmi, Maryam Mehri Dehnavi, Anand Venkat, TomofumiYuki, and Michelle M. Strout. "Extending index-array properties for data dependence analysis." LCPC'18.
  • Kazem Cheshmi, Leila Cheshmi, and Maryam Mehri Dehnavi. "Sparsity-aware storage format selection." High Performance Computing and Simulation, HPCS'18 (Best paper poster award).
  • Kazem Cheshmi, Guanzheng Xu, Saman A Zonouz, and Maryam Mehri Dehnavi. "Axb: A compiler for sparse direct solvers." Electromagnetic Field Computation, CEFC'16.
  • Kazem Cheshmi, Siamak Mohammadi, Daniel Versick, Djamshid Tavangarian, and Jelena Trajkovic. "A clustered GALS Network-on-Chip architecture with communication-aware mapping." Parallel, Distributed and Network-Based Processing, PDP'15.
  • Sebastien Le Beux, Hui Li, Ian O’Connor, Kazem Cheshmi, Xuchen Liu, Jelena Trajkovic, and Gabriela Nicolescu. "Chameleon: Channel efficient optical network-on-chip." DATE'14.
  • Kazem Cheshmi, Jelena Trajkovic, Mohammadreza Soltaniyeh, and Siamak Mohammadi, "Quota setting router architecture for quality of service in GALS network-on-chip." Rapid System Prototyping, RSP'13.

Research Experience

  • Research Intern at Microsoft Research and AI, Redmond, USA. (Jun - Sep 2019)
  • Creative Intelligence Lab Intern at Adobe Research, New York, USA. (Jun - Sep 2018)
  • Creative Technologies Lab Intern at Adobe Research, Cambridge, USA. (Jun - Sep 2017)
  • Visiting Researcher at University of Rostock, Rostock, Germany. (May - Dec 2011)

Contact

 : kazem@cs.toronto.edu
  : Github
  : Google Scholar
  : LinkedIn
  : Twitter
  : Department of Computer Science, University of Toronto, Pratt 265C, 6 King's College Rd, Toronto, ON M5S 3G4, Canada



“Capable is he who is wise - Happiness from wisdom will arise.” - Ferdowsi