projects
Research projects on high-performance data structures and large-scale data systems.
For more about these projects and the people behind them, visit the SALT Systems Lab website.
research
-
Data Systems for Efficient LLM Training and Inference GPU-native data structures and algorithms for scalable LLM training data processing, long-context inference, and serving. -
GPU-Accelerated Vector Indexes for Approximate Nearest-Neighbor Search Compact, updatable vector indexes that exploit GPU parallelism for fast approximate nearest-neighbor search. -
Scalable, Real-Time, and Information-Rich Sequence Search Over SRA Compact, exact, and incrementally updatable indexes for searching petabyte-scale sequencing data. -
Adaptive and High-performance Filters for Modern Workloads Theoretically grounded, high-performance filters (approximate membership data structures) for modern workloads. -
A Compact, Dynamic, and Distributed GPU Data Structure Library Compact, dynamic, and distributed data structures that exploit the massive parallelism of GPUs. -
Scalable Systems for Dynamic Graphs A unified system for storing, updating, and analyzing large, skewed, dynamic graphs. -
Fast and Space-Efficient In-Memory Indexes (Hash Tables, B-trees) Cache- and space-efficient in-memory hash tables and ordered indexes for scalable systems.