Tag: design-system
All the articles with the tag "design-system".
-
System Design - How Apache Iceberg Works
A beginner-friendly walkthrough of Apache Iceberg's catalog, metadata, snapshots, manifests, file pruning, commits, and the trade-offs of building reliable tables on object storage.
-
System Design - How TiDB Index Lookup Pushdown Works
A source-code walkthrough of TiDB's double-read IndexLookUp plan, how TiDB and TiKV cooperate to push local row lookups down, and the policy, hint, and affinity settings that control it.
-
AI/ML - How the LLM KV Cache Works
How the LLM key-value cache speeds up token-by-token generation, what it stores, why it consumes GPU memory, and how paged allocation helps serving systems.
-
System Design - How Unicode and UTF-8 Encoding Work
How Unicode and UTF-8 encoding work: the journey from ASCII to a universal character set, UTF-8's variable-length encoding scheme, byte structure, self-synchronization, and why it became the dominant encoding on the web.
-
System Design - How Debuggers Work
How debuggers like GDB and LLDB work under the hood: ptrace system call, INT3 software breakpoints, hardware watchpoints, single-stepping, DWARF debug info, and source code walkthrough from the GDB and Linux kernel.
-
System Design - How SQLite Works Internally
How SQLite works internally: the tokenizer, parser, bytecode compiler, virtual machine (VDBE), B-tree storage engine, pager, and WAL — a complete walkthrough of how a SQL statement travels from text to disk in the world's most deployed database.