Connection Pool Sizing: The Math That Says Fewer, Not More
Every team seems to arrive at the same configuration at some point: the database feels slow, someone searches for a
Continue readingConnection Pool Sizing: The Math That Says Fewer, Not More
Tag: performance
MHA, MQA, GQA, MLA: The KV-Cache Trade-offs Behind Every Modern Transformer
Ask someone to sketch a transformer on a whiteboard and they’ll draw the 2017 version: softmax attention over every pair
Continue readingMHA, MQA, GQA, MLA: The KV-Cache Trade-offs Behind Every Modern Transformer
Reading EXPLAIN Like a Planner: Debugging Slow Queries in PostgreSQL
# POST 2: Reading EXPLAIN A query that ran in 20 milliseconds yesterday takes 40 seconds today. Nothing in the
Continue readingReading EXPLAIN Like a Planner: Debugging Slow Queries in PostgreSQL
Five Code Gems From Quake, Redis, Linux, Python, and SQLite — and What They Teach
Some code teaches you an algorithm. Some code teaches you a way of thinking. The snippets below are in the
Continue readingFive Code Gems From Quake, Redis, Linux, Python, and SQLite — and What They Teach
Load Balancing Beyond Round Robin: Least Outstanding Requests and Consistent Hashing
Here is a failure mode that shows up in every system with more than two backends: round robin distributes requests
Continue readingLoad Balancing Beyond Round Robin: Least Outstanding Requests and Consistent Hashing
Speculative Decoding Explained: How EAGLE-3 Makes LLMs 2-3x Faster Without Changing Outputs
Autoregressive decoding is the reason large language models feel slow: every token is generated by a full forward pass, and
Zero-Downtime Schema Migrations in PostgreSQL: Expand-Contract, Lock Timeouts, and Safe ALTER TABLE
PostgreSQL takes an ACCESS EXCLUSIVE lock for most ALTER TABLE operations. That lock conflicts with everything — reads, writes, even
MVCC and VACUUM in PostgreSQL: Dead Tuples, Autovacuum Tuning, and What PG18 Changed
Every UPDATE in PostgreSQL is secretly an INSERT-plus-a-tombstone. The old row version stays on disk, fully readable to any transaction
Continue readingMVCC and VACUUM in PostgreSQL: Dead Tuples, Autovacuum Tuning, and What PG18 Changed
epoll Under the Hood: How Linux Event Loops Actually Work
Every high-concurrency network server on Linux eventually converges on the same bottleneck: not CPU, not memory, but the question of
Continue readingepoll Under the Hood: How Linux Event Loops Actually Work
Go Memory Management Under the Hood: Allocator, Stacks, Escape Analysis, and GC
Every go process you ship is quietly running a second program: the garbage collector. It shares your heap, pauses your
Continue readingGo Memory Management Under the Hood: Allocator, Stacks, Escape Analysis, and GC