Hi. I'm Yuduo Wu (吴雨多)!

I'm a Software Engineer IBM Cloud, before joining IBM, I was a Graduate Student Researcher department of electrical and computer engineering, University of California, Davis. My research area was GPU computing, and graph analytics. I was very fortunate to be advised by Professor John D. Owens and worked on an awesome project Gunrock. My Master's thesis can be found here.

Resume | Google Scholar | LinkedIn | GitHub



Recent Projects

Gunrock - High-Performance Graph Primitives on GPUs

Co-authored "Gunrock", a high-performance CUDA library for graph-processing designed specifically for GPUs. It uses a high-level, bulk-synchronous, data-centric abstraction focused on operations on a vertex or edge frontiers. Gunrock has the best performance of any programmable GPU+graph library and has better scalability with multiple GPUs on a node than any other graph library. For more details, check out our GitHub homepage and more:
[ Why-Gunrock | Poster | Publications | Documentation ]

Scene Classification based on Deep Convolutional Neural Networks (CNNs)

Collaborated with Yangzihao Wang, together we implemented a novel scene classification method which combines Convolutional Neural Network (CNN) and Spatial Pyramid Pooling to generate high-level context-aware features. We achieved better recognition rate than the state-of-the-art method on the challenging MIT Indoor-67 dataset using only the deep features trained from ImageNet. You can find the report here.

Parallel Minimum Spanning Tree (MST) Algorithm using Gunrock

Implemented parallel Boruvka's minimum spanning tree (MST) algorithm using Gunrock's data-centric programming model. Achieved up to 120X speedup against the serial CPU Boost graph library (BGL) implementation for real-world road networks, open street maps (OSMs), Delaunay graphs, and many long-diameter fully-connected synthetic datasets.

Publications

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. Gunrock: GPU Graph Analytics. CoRR, abs/1701.01170(1701.01170v1), January 2017. [ arXiv ]

Yangzihao Wang, Andrew Davidson, Yuechao Pan, Yuduo Wu, Andy Riffel, and John D. Owens. Gunrock: A High-Performance Graph Processing Library on the GPU. In Proceedings of the 21st ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming, PPoPP 2016, March 2016. Distinguished Paper Award. [ bib | DOI | http | poster | GTC talk | arXiv ]

Yuduo Wu, Yangzihao Wang, Yuechao Pan, Carl Yang, and John D. Owens. Performance Characterization of High-Level Programming Models for GPU Graph Analytics, In IEEE International Symposium on Workload Characterization (IISWC2015), October 2015. Best Paper finalist. [ bib | DOI | http | IEEE ].

Yuduo Wu, Performance Characterization of High-Level Programming Models for GPU Graph Analytics, ProQuest, September 2015. Master's thesis, University of California, Davis. [ http | PDF ]

Yuechao Pan, Yangzihao Wang, Yuduo Wu, Carl Yang, and John D. Owens. Multi-GPU Graph Analytics. CoRR, abs/1504.04804, Apr. 2015. [ bib | poster | arXiv ]

Congzhong Wu, Jie Wang, James Y.Wang, and Yuduo Wu. Study of functional verification for control endpoint of USB 3.0 In Journal of Electronic Measurement and Instrument, 2012, 26 (6)

Contact Me

LinkedIn: http://www.linkedin.com/in/yuduowu
Email: yuduowgmail.com