Carl Yang

Kemper 2250 ctcyang at ucdavis dot edu

About

I am a 5th-year PhD candidate at UC Davis being jointly advised by Prof. John Owens and Prof. Aydin Buluc (UC Berkeley, LBL). Part of my work is funded as a GSRA at LBL. I received a BASc in Electrical Engineering from University of British Columbia. My research interests include graphic processing unit (GPU) computing, parallel computing and graph processing.

Software

GraphBLAST
High-performance implementation of GraphBLAS on GPU using optimizations from Gunrock and automatic direction-optimization from Ligra. C++/CUDA header-only library.

Gunrock
High-performance graph primitives on GPU. C++/CUDA shared object library.

Service

Reviewer for the following conferences and journals:

• ACM Transactions on Parallel Computing (TOPC)

• ACM SIGGRAPH Symposium on Interactive 3D Graphics and Games (I3D)

• IEEE International Parallel & Distributed Processing Symposium (IPDPS)

Talks

Linear Algebra is the Right Way to Think About Graphs
Carl Yang, Aydin Buluc, John D. Owens
Supercomputing Doctoral Showcase, SC, November 2018
abstract | slides | poster

Single- and Multi-Machine Training with MXNet
Carl Yang, Rahul Huilgol, Haibin Lin
Internship Final Presentation, Amazon AWS, October 2018
technical report | slides | single-machine design doc | multi-machine design doc

Graph Partitioning using Bayesian Inference on GPU
Carl Yang, Steven Dalton, Maxim Naumov, Michael Garland, Aydin Buluc, John D. Owens
GPU Technology Conference, GTC, March 2018
abstract | slides | video

Design Considerations for a GraphBLAS Compliant Graph Library on Clusters of GPUs
Carl Yang, Aydin Buluc and John D. Owens
Society for Industrial and Applied Mathematics Annual Meeting, SIAM, July 2016
abstract | slides

Publications

Design Principles for Sparse Matrix Multiplication on the GPU
Carl Yang, Aydin Buluc, John D. Owens
International European Conference on Parallel and Distributed Computing, Euro-Par, August 2018
Distinguished Paper, Best Artifact
bib | paper | slides | code

Implementing Push-Pull Efficiently in GraphBLAS
Carl Yang, Aydin Buluc, John D. Owens
International Conference on Parallel Processing, ICPP, August 2018
bib | paper | slides | code

GraphBLAS C API: Ideas for future versions of the specification
Tim Mattson, Carl Yang, Scott McMillan, Aydin Buluc and Jose Moreira
High Performance Extreme Computing Conference, HPEC, September 2017
bib | paper

Multi-GPU Graph Analytics
Yuechao Pan, Yangzihao Wang, Yuduo Wu, Carl Yang and John D. Owens
International Parallel and Distributed Processing Symposium, IPDPS, May 2017
bib | paper | code

Design of the GraphBLAS API for C
Aydin Buluc, Timothy Mattson, Scott McMillan, Jose Moreira and Carl Yang
Graph Algorithm Building Blocks, GABB, May 2017
bib | paper

Gunrock: GPU Graph Analytics
Yangzihao Wang, Yuechao Pan, Andrew Davidson, Yuduo Wu, Carl Yang, Leyuan Wang, Muhammad Osama, Chenshan Yuan, Weitang Liu, Andy T. Riffel and John D. Owens
ACM Transactions on Parallel Computing, TOPC, January 2017
bib | paper | code

Mathematical Foundations of the GraphBLAS
Jeremy Kepner, Peter Aaltonen, David Bader, Aydin Buluc, Franz Franchetti, John Gilbert, Dylan Hutchison, Manoj Kumar, Andrew Lumsdaine, Henning Meyerhenke, Scott McMillan, Jose Moreira, John D. Owens, Carl Yang, Marcin Zalewski and Timothy Mattson
High Performance Extreme Computing Conference, HPEC, September 2016
bib | paper

A Comparative Study on Exact Triangle Counting Algorithms on the GPU
Leyuan Wang, Yangzihao Wang, Carl Yang and John D. Owens
1st High Performance Graph Processing Workshop, HPGP, May 2016
bib | paper

Performance Characterization of High-Level Programming Models for GPU Graph Analytics
Yuduo Wu, Yangzihao Wang, Yuechao Pan, Carl Yang and John D. Owens
Workload Characterization Symposium, IIWCS, October 2015
bib | paper

Fast Sparse Matrix and Sparse Vector Multiplication Algorithm on the GPU
Carl Yang, Yangzihao Wang and John D. Owens
Graph Algorithm Building Blocks, GABB, May 2015
bib | paper | slides | code