The Intelligent Data Infrastructure Lab at Arizona State University

Exploring Frontiers of Intelligent Data Infrastructure

Data is the gold of the 21st century, which reflects, influences, and empowers our daily lives and changes the whole world. As modern data generation continues to surge due to the explosion of social media, e-business, big data, AR/VR, and all the applications, driven by new technologies like machine learning (ML), artificial intelligence (AI), big data, cloud computing, and the Internet of Things (IoT), our infrastructure to store, manage, and process this data must continuously evolve. Our research vision is to advance the foundation of data infrastructure, including storage systems, memory systems, and database systems, by creating Resilient, Adaptive, Intelligent, Sustainable, and Efficient (RAISE) systems that not only accommodate the scale of future data but also enable seamless integration with new technologies and hardware, such as disaggregated infrastructure and emerging storage/memory devices, new applications such as large language models and scientific analysis, and urgent need for sustainable and cost-effective system designs. To achieve the research goal of RAISE, our lab mainly focuses on the system-level innovations combined with the cross-stack co-designs. We are passionate about pushing the boundaries of data infrastructure.

News

2025-08-20
[Paper]
[EMNLP 2025]
"Bit-Flip Error Resilience in LLMs: A Comprehensive Analysis and Defense Framework" accepted at EMNLP 2025
2025-07-20
[Award]
[Best Paper]
Best Student Paper Award at HPDC 2025: "LegoIndex: A Scalable and Modular Indexing Framework for Efficient Analysis of Extreme-Scale Particle Data"
2025-06-06
[Paper]
[HPDC 2025]
"LegoIndex: A Scalable and Modular Indexing Framework for Efficient Analysis of Extreme-Scale Particle Data" accepted at HPDC 2025
2025-06-06
[Paper]
[HotStorage 2025]
"Unlocking the Unusable: A Proactive Caching Framework for Reusing Partial Overlapped Data" accepted at ACM HotStorage 2025
2025-02-27
[Paper]
[arXiv]
"ELMo-Tune-V2: LLM-Assisted Full-Cycle Auto-Tuning to Optimize LSM-Based Key-Value Stores" now live on arXiv.
arXiv Link, GitHub Repo
2025-02-03
[Paper]
[SIGMOD 2025]
"SHIELD: Encrypting Persistent Data of LSM-KVS from Monolithic to Disaggregated Storage" accepted at ACM SIGMOD 2025.
PDF, GitHub Repo
2024-12-31
[Award]
[Grant]
NSF CAREER Award for proposal "Towards Disaggregated Persistent Key-Value Stores" ($730,883). Thank you NSF!
NSF Award Link, ASU News, NewsWise Article
2024-12-08
[Paper]
[IEEE Transactions on Computers]
"CPI: A Collaborative Partial Indexing Design for Large-Scale Deduplication Systems" was accepted by IEEE Transactions on Computers
2024-12-08
[Award]
[Grant]
Decoupled-LSM is awarded by NSF. Thanks to the NSF!
NSF Award Link
2024-12-08
[Paper]
[SOSP 2024]
"BIZA: Design of Self-Governing Block-Interface ZNS AFA for Endurance and Performance" has accepted by ACM SOSP 2024
  • Gallery Image
  • Gallery Image
  • Gallery Image
  • Gallery Image