Mountain View

Xia Zhao (赵夏)

Associate Researcher

Artificial Intelligence Research Center
Defense Innovation Institute
Academy of Military Science

Office: Fengtai East Street, Beijing, China
Tel/WechatID: (+86)0x448d4ebaa

I received my bachelor's degrees from School of Computer Science & Technology, Huazhong University of Science & Technology, in 2012 and my master's degrees from College of Computer, National University of Defense Technology, in 2014. I finished my Ph.D. under the advisement of Professor Lieven Eeckhout, in 2019. I am currently an Associate Researcher at the Academy of Military Science, the highest-level research institute of the People's Liberation Army.

I am also the associate director of the chip lab working on the future high-performance chip. I was selected in the Young Elite Scientists Sponsorship Program and Beijing Nova Program in 2022. My Curriculum Vitae is available here.


Education

2015-2019 PhD., Electronics and Information Systems department, Ghent University
2012-2014 M.S., College of Computer, National University of Defense Technology
2008-2012 B.S., School of Computer Science & Technology, Huazhong University of Science & Technology

Teaching


Publications

Conference Publications

  • Xia Zhao, Guangda Zhang, Lu Wang, Shiqing Zhang, Huadong Dai NearFetch: Saving Inter-Module Bandwidth in Many-Chip-Module GPUs. Accepted 31st IEEE International Symposium on High-Performance Computer Architecture (HPCA), 2025.
  • Zhongzhu Pu, Guangda Zhang, Tiejian Zhang, Chen Zhang, Youhui Zhang, Xia Zhao, ChameSC: Virtualizing Superscalar Core of a SIMD Architecture for Vector Memory Acces. In 42nd IEEE International Conference on Computer Design (ICCD), 2024.
  • Xu Zhang, Guangda Zhang, Lu Wang, Shiqing Zhang, Xia Zhao, AdCoalescer: An Adaptive Coalescer to Reduce the Inter-Module Traffic in MCM-GPUs. In 53rd International Conference on Parallel Processing (ICPP), 2024.
  • Xia Zhao, Magnus Jahre, Yuhua Tang, Guangda Zhang, Lieven Eeckhout, NUBA: Non-Uniform Bandwidth GPUs. In 28th International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS), 2023.
  • Xia Zhao, Lieven Eeckhout, Magnus Jahre, Delegated Replies: Alleviating Network Clogging in Heterogeneous Architectures. In 28th International Symposium on High-Performance Computer Architecture (HPCA), 2022.
  • Xia Zhao, Magnus Jahre, Lieven Eeckhout, Selective Replication in Memory-Side GPU Caches. In 53rd International Symposium on Microarchitecture (MICRO), 2020.
  • Xia Zhao, Magnus Jahre, Lieven Eeckhout, HSM: A Hybrid Slowdown Model for Multitasking GPUs. In 25th International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS), 2020.
  • Xia Zhao, Almutaz Adileh, Zhibin Yu, Zhiying Wang, Aamer Jaleel, Lieven Eeckhout, Adaptive Memory-Side Last-Level GPU Caching. In 46th International Symposium on Computer Architecture (ISCA), 2019.
  • Xia Zhao, Zhiying Wang, Lieven Eeckhout, Classification-Driven Search for Effective SM Partitioning in GPU Multitasking. In 32nd ACM International Conference on Supercomputing (ICS), 2018.
  • Yuxi Liu, Xia Zhao, Magnus Jahre, Zhenlin Wang, Xiaolin Wang, Yingwei Luo, Lieven Eeckhout, Get Out of the Valley: Power-Efficient Address Mapping for GPUs. In 45th International Symposium on Computer Architecture (ISCA), 2018.
  • Lu Wang, Xia Zhao, David Kaeli, Zhiying Wang, Lieven Eeckhout, Intra-Cluster Coalescing to Reduce GPU NoC Pressure. In 32nd IEEE International Parallel & Distributed Processing Symposium (IPDPS), 2018.
  • Yuxi Liu, Xia Zhao, Zhibin Yu, Zhenlin Wang, Xiaolin Wang, Yingwei Luo, Lieven Eeckhout, BACM: Barrier-Aware Cache Management for Irregular Memory-Intensive GPGPU Workloads. In 35th IEEE International Conference on Computer Design (ICCD), 2017.
  • Yuxi Liu, Xia Zhao, Zhibin Yu, Zhenlin Wang, Xiaolin Wang, Yingwei Luo, Lieven Eeckhout, BACM: Barrier-Aware Cache Management for Irregular Memory-Intensive GPGPU Workloads. In 26th IEEE International Conference on Parallel Architectures and Compilation Techniques (PACT), 2017 (Poster).
  • Xia Zhao, Sheng Ma, Yuxi Liu, Lieven Eeckhout, Zhiying Wang, A Low-Cost Conflict-Free NoC for GPGPUs. in 53rd Design Automation Conference (DAC), 2016 (Acceptance rate: 152/876=17%).
  • Xia Zhao, Sheng Ma, Chen Li, Lieven Eeckhout, Zhiying Wang, A Heterogeneous Low-Cost and Low-Latency Ring-Chain Network for GPGPUs. In 34th IEEE International Conference on Computer Design (ICCD), 2016 (Acceptance rate: 77/267=28%).

Journal Publications

  • Xia Zhao, Yuxi Liu, Almutaz Adileh, Lieven Eeckhout, LA-LLC: Inter-Core Locality-Aware Last-Level Cache To Exploit Many-to-Many Traffic in GPGPUs. IEEE Computer Architecture Letters (CAL), 2017.
  • Xia Zhao, Zhiying Wang, Lieven Eeckhout, HeteroCore GPU to Exploit TLP-Resource Diversity. IEEE Transactions on Parallel and Distributed Systems (TPDS), 2018.
  • Lu Wang, Xia Zhao, David Kaeli, Zhiying Wang, Lieven Eeckhout, Intra-Cluster Coalescing to Reduce GPU NoC Pressure. IEEE Transactions on Computers (TC), 2019.
  • Xia Zhao, Sheng Ma, Zhiying Wang, Natalie Enright Jerger, Lieven Eeckhout, CD-Xbar: A Converge-Diverge Crossbar Network for High-Performance GPUs. IEEE Transactions on Computers (TC), 2019 (Featured Paper in the September 2019).
  • Xia Zhao, Guangda Zhang, Lu Wang, Yangmei Li, Yongjun Zhang, RouteReplies: Alleviating Long Latency in Many-Chip-Module GPUs. IEEE Computer Architecture Letters (CAL), 2023.

Some Insightful/Interesting Articles and Videos



 Last modified: December 2, 2023