C
Changwan Hong
Researcher at Massachusetts Institute of Technology
Publications - 26
Citations - 484
Changwan Hong is an academic researcher from Massachusetts Institute of Technology. The author has contributed to research in topics: Speedup & Sparse matrix. The author has an hindex of 10, co-authored 25 publications receiving 279 citations. Previous affiliations of Changwan Hong include Ohio State University.
Papers
More filters
Proceedings ArticleDOI
Adaptive sparse tiling for sparse matrix multiplication
TL;DR: This paper devise an adaptive tiling strategy and apply it to enhance the performance of two primitives: SpMM (product of sparse matrix and dense matrix) and SDDMM (sampled dense-dense matrix multiplication).
Proceedings ArticleDOI
Efficient sparse-matrix multi-vector product on GPUs
Changwan Hong,Aravind Sukumaran-Rajam,Bortik Bandyopadhyay,Jinsung Kim,Sureyya Emre Kurt,Israt Nisa,Shivani Sabhlok,Ümit V. Çatalyürek,Srinivasan Parthasarathy,P. Sadayappan +9 more
TL;DR: An in-depth analysis is presented to contrast SpMV and SpMM, and a new sparse-matrix representation and computation approach suited to achieving high data-movement efficiency and effective GPU parallelization of SpMM is developed.
Journal ArticleDOI
Static and Dynamic Frequency Scaling on Multicore CPUs
Wenlei Bao,Changwan Hong,Sudheer Chunduri,Sriram Krishnamoorthy,Louis-Noël Pouchet,Fabrice Rastello,P. Sadayappan +6 more
TL;DR: This article proposes a lightweight runtime approach that can exploit the properties of the power profile specific to a processor, outperforming classical Linux governors such as powersave or on-demand for computational kernels and demonstrates that it systematically outperforms the powersave Linux governor while also improving overall performance.
Journal ArticleDOI
A sparse iteration space transformation framework for sparse tensor algebra
Ryan Senanayake,Changwan Hong,Ziheng Wang,Amalee Wilson,Stephen Chou,Shoaib Kamil,Saman Amarasinghe,Fredrik Kjolstad +7 more
TL;DR: The results show that the sparse transformations are sufficient to generate code with competitive performance to hand-optimized implementations from the literature, while generalizing to all of the tensor algebra.
Proceedings ArticleDOI
A novel data transformation and execution strategy for accelerating sparse matrix multiplication on GPUs
TL;DR: This work proposes a novel row-reordering technique to improve data locality for SpMM and SDDMM on GPUs by using a hierarchical clustering procedure optimized by locality-sensitive hashing.