Youyou Lu
Associate Professor
Works on storage systems and AI systems, including flash and persistent memory, the SuperFS file system, and vector storage and memory management for AI.
Youyou Lu's lab at Tsinghua works on distributed, cloud & edge systems. 49 papers since 2023, 17 at top venues, mostly EuroSys and OSDI. 22 PhD students and 3 master's students. 58 alumni, most often to Huawei and Alibaba. Applications close .
At a glance
- papers, 2023–2026
- 13/ year
- at T1 venues
- 35%
- PhD students
- 22
- MS students
- 3
- alumni listed
- 58
Papers by yearfrom the lab homepage, 2023–2026
- 13 papers2026 · T1 5 · collecting
- 15 papers2025 · T1 5
- 8 papers2024 · T1 5
- 13 papers2023 · T1 2
T1 venuesOtherVenues: EuroSys, OSDI, SOSP, MICRO, ICDE
Research focuspaper titles that use each topic's words
- memory15 papers
- storage12 papers
- Distributed, cloud & edge systems10 papers
- storage systems4 papers
- vector3 papers
Paper list
From the lab homepage.
202613 papers
- TOSAchieving Availability, Efficiency, and Elasticity in Stripeless Erasure-Coded Storage
- SOSP (T1)Anchor: Mitigating Shallow Disruptions with Decoupled Memory
- The USENIX Conference on File and Storage TechnologiesCost-efficient Archive Cloud Storage with Tape: Design and Deployment
- The USENIX Conference on File and Storage TechnologiesDiscard-Based Garbage Collection for Distributed Log-Structured Storage Systems in ByteDance
- SOSP (T1)Efficient GPU Multitasking with Morphable Kernels
- ACM SIGOPS Annual Technical ConferenceEfficient Training on Multiple Consumer GPUs with RoundPipe
- The USENIX Conference on File and Storage TechnologiesGPU Checkpoint/Restore Made Fast and Lightweight
- MICRO (T1)Hera: Eliminating the I/O Tax of RAID on High-Capacity SSDs
- ACM SIGMOD International Conference on Management of DataHigh-Throughput, Cost-Effective Billion-Scale Vector Search with a Single GPU
- EuroSys (T1)MesaFS: An I/O-Efficient Metadata Service for Distributed File Systems
- The USENIX Conference on File and Storage TechnologiesOdinANN: Direct Insert for Consistently Stable Performance in Billion-Scale Graph-Based Vector Search
- ICDE (T1)Switch∆: Asynchronous Metadata Updating for Distributed Storage with In-Network Data Visibility
202515 papers
- ACM Symposium on Cloud ComputingAccelerating Distributed Filesystem Metadata Service via Decoupling Directory Semantics from Metadata Indexing
- OSDI (T1)Achieving Low-Latency Graph-Based Vector Search via Aligning Best-First Search Algorithm with SSD
- NSDI (T1)Achieving Wire-Latency Storage Systems by Exploiting Hardware ACKs
- The ACM Workshop on Hot Topics in Storage and File SystemsCableCache: In-Network Request Deduplication for Key-Value Stores
- EuroSys (T1)Deft: A Scalable Tree Index for Disaggregated Memory
- CACMDesigning an Efficient Tree Index on Disaggregated Memory
- TOSEfficiently Enlarging RDMA-Attached Memory with SSD
- EuroSys (T1)Fast State Restoration in LLM Serving with HCache
- The Conference on Architectural Support for Programming Languages and Operating SystemsFrugal: Efficient and Economic Embedding Model Training with Commodity GPUs
- USENIX Annual Technical ConferenceGPreempt: GPU Preemptive Scheduling Made General and Efficient
- The Conference on Architectural Support for Programming Languages and Operating SystemsMedusa: Accelerating Serverless LLM Inference with Materialization
- The USENIX Conference on File and Storage TechnologiesShiftLock: Mitigate One-sided RDMA Lock Contention via Handover
20248 papers
- OSDI (T1)A Tale of Two Paths: Toward a Hybrid Data Plane for Efficient Far-Memory Applications
- MICRO (T1)Ares-Flash: Efficient Parallel Integer Arithmetic Operations Using NAND Flash Memory
- EuroSys (T1)Exploring the Asynchrony of Slow Memory Filesystem with EasyIO
- SOSP (T1)Fast Core Scheduling with Userspace Process Abstraction
- The Conference on Architectural Support for Programming Languages and Operating SystemsMaxEmbed: Maximizing SSD Bandwidth Utilization for Huge Embedding Models Serving
- TOSPerseid: A Secondary Indexing Mechanism for LSM-based Storage Systems
- The USENIX Conference on File and Storage TechnologiesTeRM: Extending RDMA-Attached Memory with SSD
- EuroSys (T1)Volley: Accelerating Write-Read Orders in Disaggregated Storage
202313 papers
- SIGMOD Record, March (Vol.Building Write-Optimized Tree Indexes on Disaggregated Memory
- The USENIX Conference on File and Storage TechnologiesCitron: Distributed Range Lock Management with One-sided RDMA
- TOSEfficient Crash Consistency for NVMe over PCIe and RDMA
- The Conference on Architectural Support for Programming Languages and Operating SystemsMobius: Fine Tuning Large-scale Models on Commodity GPU Servers
- Frontiers of Information Technology & Electronic EngineeringNICFS: a file system based on persistent memory and SmartNIC
- The USENIX Conference on File and Storage TechnologiesPatronus: High-Performance and Protective Remote Memory
- The International Conference on Very Large Data BasesPetPS: Supporting Huge Embedding Models with Persistent Memory
- OSDI (T1)Replicating Persistent Memory Key-Value Stores with Efficient RDMA Abstraction
- USENIX Annual Technical ConferenceRevisiting Secondary Indexing in LSM-based Storage Systems with Persistent Memory
- EuroSys (T1)RIO: Order-Preserving and CPU-Efficient Remote Storage Access
- USENIX Annual Technical ConferenceSingularFS: A Billion-Scale Distributed File System Using a Single Metadata Server
- Transactions on Embedded Computing SystemsTH-iSSD: Design and Implementation of a Generic and Reconfigurable Near-Data Processing Framework
Where alumni went58 listed, 43% to industry
Industry 10Academia 13
Most often to
- Huawei
- Alibaba
- Tencent
Also to
- Tsinghua University
- Ph.D. at Xiamen University
- Beihang University
Apply to TsinghuaFall 2027
Computer Science PhD (international applicants)
- Round
- Main
- Opens
- GRE
- See program page
- English
- Per school catalog
- Fee
- RMB 800
Not yet confirmed for this intake.
Similar labsnearest by research summary
- USUW
Michael Bedford Taylor
- Accelerators, chips & architecture
- CNSJTU
Zihan Wu
- Accelerators, chips & architecture
- CNSJTU
Chen Zhang
- Efficient training & inference
- CNSJTU
Weifeng He
- Accelerators, chips & architecture
- CNZhejiang
Kejie Huang
- Accelerators, chips & architecture
- CNSJTU
Honglan Jiang
- Accelerators, chips & architecture
Your notes
Save this lab to keep notes.
Data
Some numbers could not be verified:
- This site is shared with other labs, so counts may overlap.
- Collected
- Verified
- Papers
- The lab homepage
- Members
- Lab homepage
- Alumni
- Lab homepage
This lab shares a website with other labs. Member and alumni counts may include them.
Tiers are venue classifications, not lab rankings.
Tsinghuacloses