I believe in abstraction.
I am a PhD
student candidate at Paramathics lab, Rutgers university University of Toronto, working under supervision of Prof. Maryam Mehri Dehnavi. My research area includes optimizing numerical methods, compiler optimization, and parallel programming.
Our paper titled "Sparse Computation Data Dependence Simplification for Efficient Compiler-Generated Inspectors" is accepted in PLDI'19
Sympiler and ParSy are presented at Invited Workshop on Compiler Techniques for Sparse Tensor Algebra!.
I am honored to receive Wolfond Scholarship!.
Sympiler in the news!.
ParSy is presented in SC18, slides are available from the Sympiler website.
I joined computer science department of University of Toronto as a PhD candidate.
Our paper titled "Extending Index-Array Properties for Data Dependence Analysis" is accepted in LCPC'18
ParSy (Parallel Sympiler) is accepted at SC18! The paper is available from the following link: "ParSy: Inspection and Transformation of Sparse Matrix Computations for Parallelism"
Received the best poster paper from HPCS 2018 for paper titled: "Sparsity-Aware Storage Format Selection".
Will be in Adobe research, NY this summer as a research fellow .
I am honored to receive the Adobe Fellowship 2018 .
Presented Sympiler at Commit group at MIT.
We gave a talk on Sympiler at Lawrence Berkeley National Laboratory.
Our paper titled "Sympiler: Transforming Sparse Matrix Codes by Decoupling Symbolic Analysis" is accepted in Supercomputing 2017 .
I am honored to receive the first title in ACM SRC grand finals . I have been invited to ACM Awards banquet in San Francisco to recieve formal recognition.
I will be in Adobe Research in Cambridge, MA this summer to work on my research i.e., compiler techniques for sparse numerical methods.
I received the best poster in student research competition in CGO'17 conference.
I received NSF travel grant to present my poster in CGO'17 conference.
“Capable is he who is wise - Happiness from wisdom will arise.”
Sep 2018 - Present
PhD - Computer Science “Compiler optimization techniques for sparse numerical methods”
Sep 2009 - Sep 2012
Master of Applied Science - Computer Engineering Thesis: “Communication-Aware Task Scheduling in GALS-Based MPSoC”
Sep 2004 - Sep 2008
Bachelor of Eng - Computer Engineering Project: “Designing an Open-Source Firewall”
March 2016 - Sept 2018
Visiting Researcher Scholar Research and Teaching Assistant
Jan 2013 - Aug 2016
Research and Teaching Assistant
May 2011 - Dec 2011
Visiting Researcher Scholar
Compiler optimization and domain specific languages.
Numerical methods and sparse computations.
Parallel programming and architectures.
Formal verification methods.
“You can do anything you set your mind to.”
- Benjamin Franklin
K. Cheshmi, S. Kamil, M. M. Strout, and M. M. Dehnavi, "Sympiler: Transforming Sparse Matrix Codes by Decoupling Symbolic Analysis," In Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis (SC '17). ACM, New York, NY, USA, Article 13, 13 pages. DOI: https://doi.org/10.1145/3126908.3126936 (acceptance rate: 18%) BibTex EndNote ACM Ref
K. Cheshmi and M. M. Dehnavi, "Decoupling Symbolic from Numeric in Sparse Matrix Computations," in Proceedings of Annual IEEE/ACM International Symposium on Code Generation and Optimization. ACM, 2017 (First place in ACM SRC Grand Finals)
K. Cheshmi, S. Mohammadi, D. Versick, D. Tavangarian, and J. Trajkovic, "A clustered gals noc architecture with communication-aware mapping," in Parallel, Distributed and Network-Based Processing (PDP), 2015 23rd Euromicro International Conference on. IEEE, 2015, pp. 425-429
K. Cheshmi, J. Trajkovic, M. Soltaniyeh, and S. Mohammadi, "Quota setting router architecture for quality of service in gals noc," in Rapid System Prototyping (RSP), 2013 International Symposium on. IEEE, 2013, pp. 44-50
K. Cheshmi and M. M. Dehnavi, "Decoupling symbolic analysis from numerical factorization in sparse direct solvers," in Proceedings of Annual IEEE/ACM International Symposium on Code Generation and Optimization. ACM, 2017 (Best SRC poster award)
S. Le Beux, H. Li, I. O’Connor, K. Cheshmi, X. Liu, J. Trajkovic, and G. Nicolescu, "Chameleon: Channel efficient optical noc," in Design, Automation and Test in Europe Conference and Exhibition (DATE), 2014. IEEE, 2014, pp. 1-6
K. Cheshmi, S. Mohammadi, D. Versick, D. Tavangarian, and J. Trajkovic, "A clustered globally asynchronous locally synchronous noc communication architecture," in DIGITAL SYSTEM DESIGN(DSD) WiP session, 16TH EUROMICRO CONFERENCE ON. IEEE, 2013, pp. 12-14 (Poster)