Kai Wu   (Kyle)

I am a senior engineer at Microsoft. Previously (2021-2022), I was a researcher at ByteDance/TikTok Infrastructure System Lab. Before that, I received my Ph.D. in Electrical Engineering and Computer Sciences from University of California Merced, working with Prof. Dong Li on building efficient and scalable system supports for heterogeneous memory-based platforms. My research interests mainly focus on memory/storage systems, high performance computing (HPC), runtimes, and programming models.

Contact: kwu42 [at] ucmerced [dot] edu

Resume    /    LinkedIn    /    Google Scholar    /    Twitter    /    GitHub

News
05/2022 A paper "ByteHTAP: ByteDance's HTAP System with High Data Freshness and Strong Data Consistency" is accepted in VLDB'22.
03/2021 A paper " Optimizing Large-Scale Plasma Simulations on Persistent Memory-based Heterogeneous Memory with Effective Data Placement Across Memory Hierarchy" is accepted in ICS'21.
12/2020 A paper "ArchTM: Architecture-Aware, High Performance Transaction for Persistent Memory" is accepted in FAST'21.
10/2020 A paper "Efficient Tensor Migration and Allocation on Heterogeneous Memory Systems for Deep Learning" is accepted in HPCA'21.
09/2020 Kai will serve as shaow PC of Eurosys'20.
09/2020 Kai will serve as technical program committee of HPCC'20.
09/2020 Kai's PhD dissertation "Runtime Data Management on Non-Volatile Memory-based High Performance Systems" is selected to present in SC'20 Doctoral Showcase.
08/2020 Kai is selected to be student volunteer for SC'20.
07/2020 A paper "Exploring Non-Volatility of Non-Volatile Memory for High Performance Computing Under Failures" is accepted in CLUSTER'20.
07/2020 A paper "Ribbon: High Performance Cache Line Flushing for Persistent Memory" is accepted in PACT'20.
12/2019 A paper "Demystifying the Performance of HPC Scientific Applications on NVM-based Memory Systems" is accepted in IPDPS'20.
Selected Publications
[VLDB'22] ByteHTAP: ByteDance's HTAP System with High Data Freshness and Strong Data Consistency
Jianjun Chen, Yonghua Ding, Ye Liu, Fangshi Li, Li Zhang, Mingyi Zhang, Kui Wei, Lixun Cao, Dan Zou, Yang Liu, Lei Zhang, Rui Shi, Wei Ding, Kai Wu, Shangyu Luo, Jason Sun, Yuming Liang
[FAST'21] Architecture-Aware, High Performance Transaction for Persistent Memory
Kai Wu, Jie Ren, Ivy Peng, Dong Li
[HPCA'21] Efficient Tensor Migration and Allocation on Heterogeneous Memory Systems for Deep Learning
Jie Ren, Jiaolin Luo, Kai Wu, Mingjia Zhang, Hyeran Jeon, Dong Li
[ICS'21] Optimizing Large-Scale Plasma Simulations on Persistent Memory-based Heterogeneous Memory with Effective Data Placement Across Memory Hierarchy
Jie Ren, Jiaolin Luo, Ivy Peng, Kai Wu, Dong Li
[SC'20 Doctoral Showcase] Runtime Data Management on Non-Volatile Memory-based High Performance Systems
Kai Wu, Dong Li
[PACT'20] Ribbon: High Performance Cache Line Flushing for Persistent Memory
Kai Wu, Ivy Peng, Jie Ren, Dong Li
PDF
[IPDPS'20] Demystifying the Performance of HPC Scientific Applications on NVM-based Memory Systems
Ivy Peng, Kai Wu, Jie Ren, Dong Li, Maya Gokhale
PDF
[CLUSTER'20] Exploring Non-Volatility of Non-Volatile Memory for High Performance Computing Under Failure
Jie Ren, Kai Wu, Dong Li
PDF
[MCHPC'19] UMap: Enabling Application-driven Optimizations for Page Management
Ivy Peng, Marty McFadden, Eric Green, Keita Iwabuchi, Kai Wu, Dong Li, Roger Pearce, Maya Gokhale
PDF
[SC'18] Runtime Data Management on Non-Volatile Memory-based Heterogeneous Memory for Task-Parallel Programs
Kai Wu, Jie Ren, Dong Li
PDF
[CLUSTER'20] Exploring Non-Volatility of Non-Volatile Memory for High Performance Computing Under Failure
Jie Ren, Kai Wu, Dong Li
PDF
[ICPP'18] Modeling Application Resilience in Large Scale Parallel Execution
Kai Wu, Wenqian Dong, Qiang Guan, Nathan Debardeleben, Dong Li
PDF
[MCHPC'18] Understanding Application Recomputability without Crash Consistency in Non-Volatile Memory
Jie Ren, Kai Wu, Dong Li
PDF
[SC'17] Unimem: Runtime Data Management in Non-Volatile Memory-based Heterogeneous Main Memory
Kai Wu, Yingchao Huang, Dong Li
PDF
[CLUSTER'17] Algorithm-Directed Persistent Memory for High Performance Computing
Shuo Yang, Kai Wu, Yifan Qiao, Dong Li, Jidong Zhai
PDF
[NAS'17] Performance Evaluation and Modeling of HPC I/O on Non-Volatile Memory
Wei Liu, Kai Wu, Jialin Liu, Feng Chen, Dong Li
PDF
Professional Activities
Technical Program Committee HPCC'2022 & 2021 & 2020, Eurosys'21 (shadow PC)
Reviewer DOE SBIR/STTR, IEEE Transactions on Parallel and Distributed Systems, IEEE Access, Future Generation Computer Systems, IPDPS'21, NPC'20, CLUSTER'20, NPC'19, ICPP'19, SC'18, IPDPS'17, CLUSTER'17, HPCC'17, NAS'17
Student Volunteer SC'20, SC'19, SC'18, SC'16
Graduate Student Representative UC Merced EECS, 01/2020 - 12/2020
Awards
Conference Student Grant OSDI'20, SC'20, NVMW'20, NVMW'19, SC'19, SC'18, OSDI'18, ASPLOS'18, NVMW'18, CLUSTER'17, NVMW'17, SC'16
UC Merced Graduate Student Travel Fellowship 2018 & 2020
UC Merced Bobcat Graduate Research Fellowship 2017

© 2021 Kai Wu. Designed by Jon Barron & Limin Yang