Scholar
Wencong Xiao
Google Scholar ID: s7ky8QUAAAAJ
ByteDance
Citations & Impact
Citations
3,480
H-index
19
i10-index
24
Publications
20
Co-authors
0
Publications
Resume (English only)
Academic Achievements
- AntMan: Dynamic Scaling on GPU Clusters for Deep Learning, OSDI '20
- Distributed Graph Computation Meets Machine Learning, TPDS
- An Empirical Study on Program Failures of Deep Learning Jobs, ICSE 2020, Distinguished Paper Award!
- PRmalloc: Leveraging Predictability for Deep Learning Memory Allocation, SOSP 2019 AI Systems Workshop
- AliGraph: An Industrial Graph Neural Network Platform, SOSP 2019 AI Systems Workshop
- Analysis of Large-Scale Multi-Tenant GPU Clusters for DNN Training Workloads, ATC '19
- SeerNet: Predicting Convolutional Neural Network Feature-Map Sparsity through Low-Bit Quantization, CVPR '19
- Efficient and Effective Sparse LSTM on FPGA with Bank-Balanced Sparsity, FPGA '19
- Balanced Sparsity for Efficient DNN Inference on GPU, AAAI '19
- Scheduling CPU for GPU-based Deep Learning Jobs, SoCC '18 poster
- Gandiva: Introspective Cluster Scheduling for Deep Learning, OSDI '18
Research Experience
- Currently an AI system developer/researcher in PAI team of Alibaba Group, focusing on building a highly efficient deep learning infrastructure; previously spent over 5 years in the system research group at Microsoft Research.
Education
- Pursued Ph.D. in the system research group of Microsoft Research, supervised by Lidong Zhou from Microsoft Research and Prof. Wei Li from Beihang University.
Background
- Research interests widely spread in computer system related areas, including both traditional topics for operating systems, and modern directions with heterogeneous hardwares and new applications. Focused on providing better system support for large-scale artificial intelligent applications.
Miscellany
- Personal interests not mentioned
Co-authors