Tag: #database
Writing on GPUs, LLMs, MLOps, Kubernetes — and mindset · 114 posts
Migration Rollback and Verification — Designing a Safety Net
This post explores why rollback is hard in database migrations, and how to distinguish reversible changes from irreversible ones. It covers backups and recovery points, multi-stage verification, canary rollout, feature f
2026-06-16 · 23 min read #database#migration#rollback#verification#feature-flagOnline Schema Changes on Large Tables — gh-ost, pt-online-schema-change, and Native Online DDL
Run an ALTER against a table with hundreds of millions of rows and your service can grind to a halt. This post walks through the limits of MySQL native Online DDL, how trigger-based pt-online-schema-change and binlog-bas
2026-06-16 · 21 min read #database#mysql#postgresql#migration#gh-ostCross-Engine Database Migration — From Oracle to PostgreSQL (and Beyond)
We unpack cross-engine database migration from Oracle to PostgreSQL, starting with the real motivations of licensing and cloud. The post covers the differences between PL/SQL and PL/pgSQL, compatibility gaps in data type
2026-06-16 · 19 min read #postgresql#oracle#database#migration#ora2pgBuilding a Database Operator — Backup, Failover, and Scale as Code
A code-driven walkthrough of solving the hard parts of stateful database operations with a reconcile loop. We cover CR spec design, reconciling StatefulSet/Service/PVC, periodic backups, failover and leader-election conc
2026-06-15 · 14 min read #kubernetes#operator#database#kubebuilder#controller-runtimeBanking Ledger Design — Double-Entry Bookkeeping, Event Sourcing, and Consistency Down to the Last Cent
How to design the ledger at the heart of a banking system. We cover double-entry principles, the journal and posting data model, balance computation strategies, event sourcing, hot-account concurrency control, idempotenc
2026-06-13 · 17 min read #finance#ledger#double-entry#event-sourcing#core-bankingDuckDB in Practice — The Era of Embedded Analytics That Ends on Your Laptop
Why is DuckDB, often called the SQLite of OLAP, so loved? We start from the architecture and cover direct Parquet querying, hands-on examples of window functions, PIVOT, and ASOF JOIN, zero-copy integration with pandas,
2026-06-12 · 14 min read #duckdb#database#olap#analytics#parquetDatabase Engines 2026 Deep-Dive — Postgres Won the API, ClickHouse and DuckDB Won Analytics
PostgreSQL 17/18, MySQL 9, MariaDB 12, SQLite + libSQL + Turso, DuckDB 1.x, ClickHouse, StarRocks, Doris, Druid, Pinot, TiDB, CockroachDB 25, YugabyteDB, Cassandra 5, ScyllaDB, Aerospike, Neon, Supabase, PlanetScale, Xat
2026-05-16 · 17 min read #database#postgresql#mysql#clickhouse#duckdbDB Migration Tools 2026 — Atlas / Flyway / Liquibase / Bytebase / pgroll / Squawk / gh-ost Deep Dive
The 2026 landscape of database schema migration tools. Atlas declarative model, the Flyway 11 vs Liquibase 4.30 camps, Sqitch dependency graphs, Bytebase 3 DevOps workflow, pgroll zero-downtime expand-contract, Squawk li
2026-05-16 · 25 min read #database#migration#schema-management#atlas#flywayDatabase Admin GUI Tools 2026 — TablePlus / Beekeeper / DBeaver / DataGrip / pgAdmin / MongoDB Compass / RedisInsight / MySQL Workbench / Azure Data Studio / Querious / Postico / Outerbase / Supabase Studio Deep Dive
A 2026 map of database admin GUI tools. The general-purpose camp (TablePlus, Beekeeper Studio, DBeaver, DataGrip), the official tools (pgAdmin 4, MongoDB Compass, RedisInsight, MySQL Workbench, Azure Data Studio), the Ma
2026-05-16 · 20 min read #database#gui#admin#tableplus#beekeeper-studioTime-Series Databases in 2026 — TimescaleDB / InfluxDB 3 / QuestDB / ClickHouse / VictoriaMetrics Deep Dive
The 2026 time-series database map — TimescaleDB (the 1 Postgres extension), InfluxDB 3 (reborn on Apache DataFusion + Arrow), QuestDB (SQL plus fast ingest), ClickHouse (a column store with dominant analytical performanc
2026-05-15 · 18 min read #time-series#database#tsdb#timescaledb#influxdbDistributed SQL / NewSQL 2026 — CockroachDB / TiDB / YugabyteDB / Spanner / Aurora DSQL / Neon Deep Dive
The 2026 distributed SQL landscape no longer fits under the single word "NewSQL" that defined the early 2020s. CockroachDB 24 has settled into its position eighteen months after its license change, TiDB 8 has dragged the
2026-05-15 · 18 min read #database#distributed-sql#newsql#cockroachdb#tidbTypeScript ORMs and Query Builders 2026 — Drizzle vs Kysely vs Prisma vs postgres.js Deep Dive (How Close to SQL?) (english)
In 2026, TypeScript database access is no longer a 'Prisma or raw SQL' binary. Drizzle has claimed the headless-ORM seat with the closest-to-SQL ergonomics, Kysely has won over raw-SQL teams as a schemaless type-safe que
2026-05-14 · 20 min read #typescript#orm#drizzle#kysely#prismaPostgreSQL Production Operations Complete Guide 2025: Extensions, Tuning, Monitoring, HA
Everything about running PostgreSQL in production! Essential extensions (pgstatstatements/pgvector/PostGIS/TimescaleDB), performance tuning (sharedbuffers/workmem/effectivecachesize), VACUUM/Autovacuum management, monito
2026-04-15 · 7 min read #postgresql#database#production#extensions#performancePostgreSQL Internals Deep Dive — MVCC, VACUUM, WAL, Query Planner, Index, Partitioning, pgvector (2025)
Why did PostgreSQL become "the second most powerful DB after Oracle"? MVCC tuple versioning and the VACUUM burden, WAL and streaming replication, how the Query Planner builds execution plans, choosing among B-Tree/GIN/BR
2026-04-15 · 14 min read #postgresql#database#mvcc#query-optimization#infraLSM-Tree Deep Dive — RocksDB, LevelDB, Compaction, Bloom Filter, Write Amplification (2025)
Everything about LSM-Tree — the 1996 O'Neil paper, Memtable and SSTable structure, WAL and flush, Leveled vs Tiered vs Universal compaction, Bloom filters and read path, the RUM conjecture (Read-Update-Memory), RocksDB c
2026-04-15 · 13 min read #lsm-tree#rocksdb#leveldb#cassandra#scyllaDuckDB Internals Deep Dive — Embedded OLAP, Vectorized Execution, Morsel-Driven Parallelism, and Storage Format (2025)
DuckDB — the analytical engine that replaces Pandas and brings a ClickHouse-class OLAP database into your process. This post dissects DuckDB from the ground up: the journey from Mark Raasveldt and Hannes Mühleisens 2018
2026-04-15 · 24 min read #duckdb#olap#database#columnar#vectorizationDistributed Databases 2025 — CockroachDB, Spanner, TiDB, Yugabyte, Aurora DSQL, Neon, PlanetScale, Turso, D1 (S7 E3)
The "just use one database" era is over. In 2024–2025 distributed SQL (CockroachDB, Spanner, TiDB, Yugabyte), serverless Postgres (Neon, Supabase, PlanetScale, Aurora DSQL), and edge SQLite (Turso/libSQL, Cloudflare D1)
2026-04-15 · 11 min read #database#distributed-sql#postgresql#cockroachdb#spannerDB Sharding & Partitioning Complete Guide 2025: Horizontal Scaling, Range/Hash/Directory, Resharding, Vitess
Everything about database sharding! Horizontal vs vertical partitioning, shard key selection, Range/Hash/Directory/Geographic sharding, hot spot problems, resharding, distributed transactions, Vitess/Citus/CockroachDB co
2026-04-15 · 15 min read #database#sharding#partitioning#horizontal-scaling#vitessDB Index Master Guide 2025: B-Tree, Hash, GIN, BRIN, Composite Indexes, Query Plan Analysis
Everything about DB indexes. B-Tree internals, comparison of Hash/GIN/GiST/BRIN, composite index design, covering/partial/expression indexes, reading EXPLAIN ANALYZE, 10 reasons an index is skipped, real-world tuning cas
2026-04-15 · 10 min read #database#indexing#btree#hash#ginDatabase Query Optimizer Complete Guide 2025: Volcano, Cascades, Cost-Based, Cardinality Estimation — PostgreSQL/MySQL In-Depth Analysis
A single SQL statement can turn into hundreds of different execution plans. From Volcano to Cascades, how cost-based optimizers work, the pitfalls of cardinality estimation, and the internals of PostgreSQL and MySQL — a
2026-04-15 · 24 min read #database#query-optimizer#volcano#cascades#cost-based