Improving storage efficiency in Magic Pocket, Dropbox's immutable blob store  💾Flash Storage

Two Studies in Compiler Optimisations  📐E-graphs

Simulating Catalog and Table Conflicts in Iceberg  🧊Iceberg

Beyond Cosine Similarity  🔍Vector Search Algorithms

Simulating Queueing  ⚙️Adaptive Execution

Designing a SIMD Algorithm from Scratch  ⚡SIMD Optimization

Notes on Writing WASM  📦WASM

How HNSW Makes Vector Search Fast  🗂️Vector Indexes

An Interactive Intro to CRDTs  🔀CRDTs

Testing Datadog Explain Plan Visualizer With Oracle Execution Plans  🧠Query Planners

Exposing More Parallelism Is the Hidden Reason Why Some Vectorized Loops Are Faster - Not Vectorization per se  ⚡SIMD Optimization

Building Index-Backed Query Plans in DataFusion  🧠Query Planners

ann_research  📏ANN Benchmarks

Deep Dive: How StarRocks Built a High-Performance Vectorized Engine  ⚡SIMD Vectorization

Single Store Vector Search Index: Architecture and Memory Efficiency  🗂️Vector Indexes

Firefly: Illuminating the path to nanosecond-level clock sync in the data center  ⚙️Mechanical Sympathy

ByteHouse: A Cloud-Native OLAP Engine with Incremental Computation and Multi-Modal Retrieval  📦In-process Databases

Lessons from BF-Tree: Building a Concurrent Larger-Than-Memory Index in Rust  🗃️Database Storage

ahead-of-time wasm gc in wastrel  📦WASM

Scaling Laws for Embedding Dimension in Information Retrieval  📌Embedding Retrieval