Junyi Liu

Principal Researcher

Publications

View by:

- OasisKV: Scaling In-Decode KV Cache Beyond HBM with Lookahead Sparse Prefetching
  
  Can Xiao, Sukmin Cho, Junbong We, Zhixiong Niu, Jianyi Cheng, Yiren Zhao, Youngjin Kwon, Yongqiang Xiong, Rui Ma, Junyi Liu
  
  August 2026
  
  arXiv | August 2026
  
  Publication
- MemExplorer: Navigating the Heterogeneous Memory Design Space for Agentic Inference NPUs
  
  Haoran Wu, Zeyu Cao, Yao Lai, Binglei Lou, Jiayi Nie, Can Xiao, T. Adeniran, Przemyslaw Forys, Kauser Johar, Catriona R Wright, Junyi Liu, Kai Shi, Nicholas D. Lane, R. Antonova, Jianyi Cheng, Timothy Jones, Aaron Zhao, Robert Mullins
  
  April 2026
  
  arXiv | April 2026
  
  Publication
- Mosaic: Breaking the Optics versus Copper Trade-off with a Wide-and-Slow Architecture and MicroLEDs
  
  Kaoutar Benyahya, Ariel Gomez Diaz, Junyi Liu, Vassily Lyutsarev, Marianna Pantouvaki, Kai Shi, Shawn Yohanes Siew, Hitesh Ballani, Thomas Burridge, Daniel Cletheroe, Thomas Karagiannis, Brian Robertson, Ant Rowstron, Mengyang Yang, Arash Behziz, Jamie Gaudette, Paolo Costa
  
  The ACM SIGCOMM Conference on Data Communication (SIGCOMM) 2025 | September 2025
  
  SIGCOMM 2025 Best Paper Award
  
  PDF Video
- Good things come in small packages: Should we build AI clusters with Lite-GPUs?
  
  Burcu Canakci, Junyi Liu, Xingbo Wu, Nathanael Cheriere, Paolo Costa, Sergey Legtchenko, Dushyanth Narayanan, Ant Rowstron
  
  The ACM SIGOPS 20th Workshop on Hot Topics in Operating Systems | May 2025
  
  Publication
- Storage Class Memory is Dead, All Hail Managed-Retention Memory: Rethinking Memory for the AI Era
  
  Sergey Legtchenko, Ioan Stefanovici, Richard Black, Ant Rowstron, Junyi Liu, Paolo Costa, Burcu Canakci, Dushyanth Narayanan, Xingbo Wu
  
  The ACM SIGOPS 20th Workshop on Hot Topics in Operating Systems | May 2025
  
  Publication
- Honeycomb: Ordered Key-Value Store Acceleration on an FPGA-Based SmartNIC
  
  Junyi Liu, Aleksandar Dragojevic, Shane Fleming, Antonios Katsarakis, Dario Korolija, Igor Zablotchi, Ho-Cheung Ng, Anuj Kalia, Miguel Castro
  
  IEEE Transactions on Computers | December 2023, Vol 73: pp. 857-871
  
  Publication Publication DOI Project

- OasisKV: Scaling In-Decode KV Cache Beyond HBM with Lookahead Sparse Prefetching
  
  Can Xiao, Sukmin Cho, Junbong We, Zhixiong Niu, Jianyi Cheng, Yiren Zhao, Youngjin Kwon, Yongqiang Xiong, Rui Ma, Junyi Liu
  
  August 2026
  
  arXiv | August 2026
  
  Publication
- Storage Class Memory is Dead, All Hail Managed-Retention Memory: Rethinking Memory for the AI Era
  
  Sergey Legtchenko, Ioan Stefanovici, Richard Black, Ant Rowstron, Junyi Liu, Paolo Costa, Burcu Canakci, Dushyanth Narayanan, Xingbo Wu
  
  The ACM SIGOPS 20th Workshop on Hot Topics in Operating Systems | May 2025
  
  Publication
- Good things come in small packages: Should we build AI clusters with Lite-GPUs?
  
  Burcu Canakci, Junyi Liu, Xingbo Wu, Nathanael Cheriere, Paolo Costa, Sergey Legtchenko, Dushyanth Narayanan, Ant Rowstron
  
  The ACM SIGOPS 20th Workshop on Hot Topics in Operating Systems | May 2025
  
  Publication
- MemExplorer: Navigating the Heterogeneous Memory Design Space for Agentic Inference NPUs
  
  Haoran Wu, Zeyu Cao, Yao Lai, Binglei Lou, Jiayi Nie, Can Xiao, T. Adeniran, Przemyslaw Forys, Kauser Johar, Catriona R Wright, Junyi Liu, Kai Shi, Nicholas D. Lane, R. Antonova, Jianyi Cheng, Timothy Jones, Aaron Zhao, Robert Mullins
  
  April 2026
  
  arXiv | April 2026
  
  Publication
- Good things come in small packages: Should we build AI clusters with Lite-GPUs?
  
  Burcu Canakci, Junyi Liu, Xingbo Wu, Nathanael Cheriere, Paolo Costa, Sergey Legtchenko, Dushyanth Narayanan, Ant Rowstron
  
  The ACM SIGOPS 20th Workshop on Hot Topics in Operating Systems | May 2025
  
  Publication
- Honeycomb: Ordered Key-Value Store Acceleration on an FPGA-Based SmartNIC
  
  Junyi Liu, Aleksandar Dragojevic, Shane Fleming, Antonios Katsarakis, Dario Korolija, Igor Zablotchi, Ho-Cheung Ng, Anuj Kalia, Miguel Castro
  
  IEEE Transactions on Computers | December 2023, Vol 73: pp. 857-871
  
  Publication Publication DOI Project
- MemExplorer: Navigating the Heterogeneous Memory Design Space for Agentic Inference NPUs
  
  Haoran Wu, Zeyu Cao, Yao Lai, Binglei Lou, Jiayi Nie, Can Xiao, T. Adeniran, Przemyslaw Forys, Kauser Johar, Catriona R Wright, Junyi Liu, Kai Shi, Nicholas D. Lane, R. Antonova, Jianyi Cheng, Timothy Jones, Aaron Zhao, Robert Mullins
  
  April 2026
  
  arXiv | April 2026
  
  Publication
- Mosaic: Breaking the Optics versus Copper Trade-off with a Wide-and-Slow Architecture and MicroLEDs
  
  Kaoutar Benyahya, Ariel Gomez Diaz, Junyi Liu, Vassily Lyutsarev, Marianna Pantouvaki, Kai Shi, Shawn Yohanes Siew, Hitesh Ballani, Thomas Burridge, Daniel Cletheroe, Thomas Karagiannis, Brian Robertson, Ant Rowstron, Mengyang Yang, Arash Behziz, Jamie Gaudette, Paolo Costa
  
  The ACM SIGCOMM Conference on Data Communication (SIGCOMM) 2025 | September 2025
  
  SIGCOMM 2025 Best Paper Award
  
  PDF Video
- Storage Class Memory is Dead, All Hail Managed-Retention Memory: Rethinking Memory for the AI Era
  
  Sergey Legtchenko, Ioan Stefanovici, Richard Black, Ant Rowstron, Junyi Liu, Paolo Costa, Burcu Canakci, Dushyanth Narayanan, Xingbo Wu
  
  The ACM SIGOPS 20th Workshop on Hot Topics in Operating Systems | May 2025
  
  Publication
- Good things come in small packages: Should we build AI clusters with Lite-GPUs?
  
  Burcu Canakci, Junyi Liu, Xingbo Wu, Nathanael Cheriere, Paolo Costa, Sergey Legtchenko, Dushyanth Narayanan, Ant Rowstron
  
  The ACM SIGOPS 20th Workshop on Hot Topics in Operating Systems | May 2025
  
  Publication
- Honeycomb: Ordered Key-Value Store Acceleration on an FPGA-Based SmartNIC
  
  Junyi Liu, Aleksandar Dragojevic, Shane Fleming, Antonios Katsarakis, Dario Korolija, Igor Zablotchi, Ho-Cheung Ng, Anuj Kalia, Miguel Castro
  
  IEEE Transactions on Computers | December 2023, Vol 73: pp. 857-871
  
  Publication Publication DOI Project

- OasisKV: Scaling In-Decode KV Cache Beyond HBM with Lookahead Sparse Prefetching
  
  Can Xiao, Sukmin Cho, Junbong We, Zhixiong Niu, Jianyi Cheng, Yiren Zhao, Youngjin Kwon, Yongqiang Xiong, Rui Ma, Junyi Liu
  
  August 2026
  
  arXiv | August 2026
  
  Publication
- MemExplorer: Navigating the Heterogeneous Memory Design Space for Agentic Inference NPUs
  
  Haoran Wu, Zeyu Cao, Yao Lai, Binglei Lou, Jiayi Nie, Can Xiao, T. Adeniran, Przemyslaw Forys, Kauser Johar, Catriona R Wright, Junyi Liu, Kai Shi, Nicholas D. Lane, R. Antonova, Jianyi Cheng, Timothy Jones, Aaron Zhao, Robert Mullins
  
  April 2026
  
  arXiv | April 2026
  
  Publication
- Honeycomb: Ordered Key-Value Store Acceleration on an FPGA-Based SmartNIC
  
  Junyi Liu, Aleksandar Dragojevic, Shane Fleming, Antonios Katsarakis, Dario Korolija, Igor Zablotchi, Ho-Cheung Ng, Anuj Kalia, Miguel Castro
  
  IEEE Transactions on Computers | December 2023, Vol 73: pp. 857-871
  
  Publication Publication DOI Project
- Mosaic: Breaking the Optics versus Copper Trade-off with a Wide-and-Slow Architecture and MicroLEDs
  
  Kaoutar Benyahya, Ariel Gomez Diaz, Junyi Liu, Vassily Lyutsarev, Marianna Pantouvaki, Kai Shi, Shawn Yohanes Siew, Hitesh Ballani, Thomas Burridge, Daniel Cletheroe, Thomas Karagiannis, Brian Robertson, Ant Rowstron, Mengyang Yang, Arash Behziz, Jamie Gaudette, Paolo Costa
  
  The ACM SIGCOMM Conference on Data Communication (SIGCOMM) 2025 | September 2025
  
  SIGCOMM 2025 Best Paper Award
  
  PDF Video
- Storage Class Memory is Dead, All Hail Managed-Retention Memory: Rethinking Memory for the AI Era
  
  Sergey Legtchenko, Ioan Stefanovici, Richard Black, Ant Rowstron, Junyi Liu, Paolo Costa, Burcu Canakci, Dushyanth Narayanan, Xingbo Wu
  
  The ACM SIGOPS 20th Workshop on Hot Topics in Operating Systems | May 2025
  
  Publication
- Good things come in small packages: Should we build AI clusters with Lite-GPUs?
  
  Burcu Canakci, Junyi Liu, Xingbo Wu, Nathanael Cheriere, Paolo Costa, Sergey Legtchenko, Dushyanth Narayanan, Ant Rowstron
  
  The ACM SIGOPS 20th Workshop on Hot Topics in Operating Systems | May 2025
  
  Publication

Junyi Liu

Publications

OasisKV: Scaling In-Decode KV Cache Beyond HBM with Lookahead Sparse Prefetching

MemExplorer: Navigating the Heterogeneous Memory Design Space for Agentic Inference NPUs

Mosaic: Breaking the Optics versus Copper Trade-off with a Wide-and-Slow Architecture and MicroLEDs

Good things come in small packages: Should we build AI clusters with Lite-GPUs?

Storage Class Memory is Dead, All Hail Managed-Retention Memory: Rethinking Memory for the AI Era

Honeycomb: Ordered Key-Value Store Acceleration on an FPGA-Based SmartNIC

OasisKV: Scaling In-Decode KV Cache Beyond HBM with Lookahead Sparse Prefetching

Storage Class Memory is Dead, All Hail Managed-Retention Memory: Rethinking Memory for the AI Era

Good things come in small packages: Should we build AI clusters with Lite-GPUs?

MemExplorer: Navigating the Heterogeneous Memory Design Space for Agentic Inference NPUs

Good things come in small packages: Should we build AI clusters with Lite-GPUs?

Honeycomb: Ordered Key-Value Store Acceleration on an FPGA-Based SmartNIC

MemExplorer: Navigating the Heterogeneous Memory Design Space for Agentic Inference NPUs

Mosaic: Breaking the Optics versus Copper Trade-off with a Wide-and-Slow Architecture and MicroLEDs

Storage Class Memory is Dead, All Hail Managed-Retention Memory: Rethinking Memory for the AI Era

Good things come in small packages: Should we build AI clusters with Lite-GPUs?

Honeycomb: Ordered Key-Value Store Acceleration on an FPGA-Based SmartNIC

OasisKV: Scaling In-Decode KV Cache Beyond HBM with Lookahead Sparse Prefetching

MemExplorer: Navigating the Heterogeneous Memory Design Space for Agentic Inference NPUs

Honeycomb: Ordered Key-Value Store Acceleration on an FPGA-Based SmartNIC

Mosaic: Breaking the Optics versus Copper Trade-off with a Wide-and-Slow Architecture and MicroLEDs

Storage Class Memory is Dead, All Hail Managed-Retention Memory: Rethinking Memory for the AI Era

Good things come in small packages: Should we build AI clusters with Lite-GPUs?

Contact Junyi Liu

Microsoft Research Lab – Cambridge