Tag: design-database
All the articles with the tag "design-database".
-
System Design - How Vector Databases Work
How vector databases work: embedding vectors, similarity search, indexing with HNSW and IVF, distance metrics, quantization, and a look at how Milvus organizes its storage engine.
-
System Design - How Database Query Optimization Works
How database query optimizers work: parsing SQL into trees, generating candidate plans, cost estimation, join ordering with dynamic programming and genetic algorithms, and a source code walkthrough of the PostgreSQL optimizer.
-
System Design - How Bloom Filters Work
How Bloom filters work: the probabilistic data structure behind fast membership testing in databases, with a source code walkthrough of LevelDB and RocksDB implementations.
-
System Design - How B+ Trees Power Every Database Index
How B+ trees work: node structure, search, insertion with splits, deletion with merges, disk I/O optimization, and why virtually every relational database uses them for indexing.
-
System Design - How Write-Ahead Logging (WAL) Works
How Write-Ahead Logging (WAL) works: the core protocol behind database durability and crash recovery. Covers the WAL principle, record structure, buffer management, checkpoints, crash recovery, and a source code walkthrough from PostgreSQL.
-
System Design - How LSM Trees Work (The Engine Behind RocksDB, LevelDB, and Cassandra)
How Log-Structured Merge Trees (LSM Trees) work: the write-optimized data structure behind RocksDB, LevelDB, Cassandra, and TiKV. Covers memtable, WAL, SSTables, compaction strategies, bloom filters, and a source code walkthrough from the facebook/rocksdb repository.