I’m interested in fault-tolerant distributed and concurrent algorithms and systems, theory and practice.

Scalable Byzantine Fault Tolerance

• Tame the Wild with Byzantine Linearizability: Reliable Broadcast, Snapshots, and Asset Transfer (DISC’21)
• All You Need is DAG (PODC’21), video
• Byzantine Agreement with Less Communication: Recent Advances (SIGACT News 2021)
• Not a COINcidence: Sub-Quadratic Asynchronous Byzantine Agreement WHP (DISC’20), video
• Expected Linear Round Synchronization: The Missing Link for Linear Byzantine SMR (DISC’20), video
• FairLedger: A Fair Blockchain Protocol for Financial Institutions (OPODIS’19 Best Paper Award)

Concurrent Data Structures & Memory Transactions

• Oak: A Scalable Off-Heap Allocated Key-Value Map (PPoPP’20); Oak repository
• KiWi: A Key-Value Map for Scalable Real-Time Analytics (PPoPP’17); KiWi code
• Transactional Data Structure Libraries (PLDI’16 Distinguished Paper); Java transactional libraries code
• CSR: Core Surprise Removal in Commodity Operating Systems; uses HTM (ASPLOS’16)
• Multi-versioning in Transactional Memory (Chapter in Transactional Memory)

Data Sketches

• Fast Concurrent Data Sketches  (PPoPP’20); code (apache incubator)
• Intermediate Value Linearizability: A Quantitative Correctness Criterion (DISC’20 Best Student Paper) 

Security

• WatchIT: Who Watches Your IT Guy? (SOSP’17)

Distributed Storage

Theory

• Space Bounds: Integrated Bounds for Disintegrated Storage (DISC’18), slides
• Space Bounds for Reliable Storage: Fundamental Limits of Coding (PODC’16)
• Dynamic Reconfiguration: Abstraction and Optimal Asynchronous Solution (DISC’17)

Systems

• EvenDB: Optimizing Key-Value Storage for Spatial Locality (EuroSys’20)
• Transactions over HBase: Omid, Reloaded: Scalable and Highly-Available Transaction Processing (FAST’17) Apache Phoenix Omid
  Taking Omid to the Clouds: Fast, Scalable Transactions for Real-Time Cloud Analytics (PVLDB’18)
• HBase memory store improvement: Accordion: Better Memory Organization for LSM Key-Value Stores (PVLDB’18)
• cLSM: Improving concurrency in KV stores (LevelDB, RocksDB): Scaling Concurrent Log-Structured Data Stores (EuroSys’15); cLSM implementation in RocksDB
• ZooKeeper in multiple data centers: Modular Composition of Coordination Services (Usenix ATC’16); ZooNet code