Best of Database — October 2025

  1. 1
    Article
    Avatar of gcgitconnected·31w

    How to Scale Like a Senior Engineer (Servers, DBs, LBs, SPOFs)

    Scaling systems is about solving problems incrementally, not jumping to complex solutions. Start with understanding single server limitations, identify bottlenecks (CPU, memory, disk I/O), then make informed decisions about vertical vs horizontal scaling. Database optimization is often the real bottleneck, not application servers. Load balancers require careful algorithm selection and configuration. Even redundant architectures have single points of failure that need identification and mitigation. The key is adding complexity only when necessary and understanding the tradeoffs at each step.

  2. 2
    Article
    Avatar of hnHacker News·31w

    Your data model is your destiny

    A product's data model—the core concepts and objects it prioritizes—determines whether new features create compounding advantages or just add to a feature list. Companies like Slack (persistent channels), Notion (blocks), Figma (shared canvas), and Rippling (employee records) succeeded by choosing non-obvious data models that became impossible for competitors to replicate without rebuilding from scratch. As AI commoditizes code execution, the data model becomes the primary moat. Horizontal tools innovate on how products are built, while vertical tools succeed by elevating the right domain objects. The key is identifying the atomic unit of work in your domain and ensuring every new feature strengthens that central concept.

  3. 3
    Article
    Avatar of bytebytegoByteByteGo·32w

    How Flipkart Built a Highly Available MySQL Cluster for 150+ Million Users

    Flipkart built Altair, an internally managed MySQL service that maintains high availability for 150+ million daily users through automated failover and primary-replica architecture. The system uses a three-layered monitoring approach (agent, monitor, orchestrator) to detect failures, prevent false positives, and execute failovers with minimal data loss. Altair prioritizes write availability over strong consistency using asynchronous replication, implements DNS-based service discovery for seamless failovers, and includes multiple safeguards against split-brain scenarios. The design balances operational simplicity with reliability, achieving near five-nines availability while managing thousands of database clusters across Flipkart's microservices infrastructure.

  4. 4
    Article
    Avatar of hnHacker News·31w

    From Millions to Billions

    Geocodio migrated their request logging system from MariaDB with deprecated TokuDB to ClickHouse Cloud after hitting performance issues at billions of monthly requests. The solution involved introducing Kafka for event streaming and Vector for batch processing, moving from individual row inserts to batched inserts of 30k-50k records. The migration strategy used feature flags to run both systems in parallel, enabling zero-downtime validation before fully switching over.

  5. 5
    Article
    Avatar of newstackThe New Stack·32w

    Why We Created Turso, a Rust-Based Rewrite of SQLite

    Turso is a Rust-based rewrite of SQLite designed to address modern development needs while maintaining compatibility. SQLite's single-writer architecture, synchronous design, and limited extensibility create bottlenecks for high-throughput writes, real-time applications, and modern features like vector search and CDC. Turso introduces asynchronous I/O, concurrent writes through MVCC, native encryption, vector search, CDC support, and live materialized views. The rewrite enables SQLite-like simplicity with architectural improvements for edge computing, AI agents, and streaming analytics while fostering an open contribution model with over 150 contributors.

  6. 6
    Article
    Avatar of elixirstatusElixirStatus·30w

    You Don't Need an ORM

    Explores an alternative approach to database access that embraces raw SQL instead of abstracting it away with ORMs. Introduces Squirrel, a library for Gleam that uses code generation from SQL queries to provide type-safety and good developer experience without sacrificing performance. Demonstrates how developers can write plain SQL while maintaining the benefits of static typing in functional languages, avoiding common ORM pitfalls like N+1 query problems.

  7. 7
    Article
    Avatar of hnHacker News·31w

    Why Is SQLite Coded In C

    SQLite remains implemented in C because it provides optimal performance, universal compatibility across platforms and languages, minimal runtime dependencies, and stability. C enables SQLite to be called from any programming language, requires only basic standard library functions, and avoids the complexity of modern language runtimes. While safe languages like Rust are acknowledged as potential future options, the SQLite team prioritizes mature, well-understood languages and notes that recoding would likely introduce more bugs than it would fix. The article addresses common questions about why SQLite wasn't built with object-oriented languages or modern safe alternatives.

  8. 8
    Article
    Avatar of mongoMongoDB·33w

    The 10 Skills I Was Missing as a MongoDB User

    A developer shares lessons learned from transitioning from relational databases to MongoDB, covering common mistakes like improper data modeling (over-normalizing then over-embedding), ineffective indexing strategies, and reactive monitoring. The piece walks through how understanding document modeling patterns, query optimization with explain plans, aggregation pipelines, and proactive monitoring transformed their MongoDB experience from frustrating to efficient. MongoDB's new Skill Badges program is highlighted as a structured learning path that covers these fundamental skills.

  9. 9
    Article
    Avatar of jetbrainsJetBrains·33w

    DataGrip Is Now Free for Non-Commercial Use

    JetBrains DataGrip, a cross-platform database IDE, is now free for non-commercial use including learning, hobby projects, open-source development, and content creation. The change follows similar moves for RustRover, CLion, Rider, WebStorm, and RubyMine. All commercial features remain available in the free version, including AI-powered code completion, multi-database support, and Git integration. Commercial users must still purchase licenses. The free license lasts one year with automatic renewal and requires anonymous telemetry sharing.

  10. 10
    Article
    Avatar of systemdesignnewsSystem Design Newsletter·32w

    System Design Interview: Design Spotify

    A comprehensive guide to designing a music streaming platform like Spotify for system design interviews. Covers architecture components including blob storage for audio files, SQL databases for metadata, CDN integration, and API design. Explores capacity planning for 500K users and 30M songs, read/write workflows, and scaling strategies like database replication, sharding, and horizontal scaling. Includes practical considerations for audio delivery, caching, reliability patterns, and monitoring metrics.

  11. 11
    Article
    Avatar of bytebytegoByteByteGo·33w

    How Airbnb Runs Distributed Databases on Kubernetes at Scale

    Airbnb deployed distributed SQL databases across multiple Kubernetes clusters, each mapped to a different AWS Availability Zone, to achieve high availability and fault tolerance. They built custom Kubernetes operators to safely manage stateful workloads, coordinate node replacements, and maintain quorum during failures. Using AWS EBS for persistent storage, PVCs for volume management, and techniques like replica reads and stale reads, they mitigated latency issues while maintaining consistency. Their largest production cluster handles 3 million queries per second across 150 nodes with 300TB of data, achieving 99.95% availability through careful sequencing of upgrades, canary deployments, and overprovisioning for resilience.

  12. 12
    Article
    Avatar of systemdesigncodexSystem Design Codex·30w

    How Paypal Built JunoDB?

    PayPal open-sourced JunoDB, a distributed key-value store handling 350 billion daily requests with 99.9999% availability. Built in Go for multi-core support and CPU-intensive operations like encryption, JunoDB uses a proxy-based architecture with RocksDB for storage. Unlike Redis's single-threaded, memory-bound design, JunoDB is optimized for CPU-bound workloads. Common use cases include caching with flexible TTLs, implementing idempotency for payment processing, tracking counters for rate limiting, and bridging replication latency between data centers in active-active database configurations.

  13. 13
    Article
    Avatar of cratedbCrateDB·33w

    New Release: CrateDB 6.0

    CrateDB 6.0 upgrades to Lucene 10.2, delivering faster query performance through optimized COUNT operations, mixed joins, and improved NUMERIC type handling. The release strengthens PostgreSQL compatibility with aligned scalar functions and metadata improvements, while introducing new monitoring tools like cluster_health and expanded segments table. Dynamic column policy now supports nested arrays for easier JSON and IoT data ingestion. The Lucene upgrade requires reindexing from older versions but establishes groundwork for future AI features including vector search and k-means clustering.

  14. 14
    Article
    Avatar of gnjkg5k0npw7asw1zhhu6Suraj Shetty·29w

    Every Developer Should Know What Their DBA Knows

    A comprehensive roadmap covering essential PostgreSQL database administration knowledge that developers should understand. Topics include database fundamentals, PostgreSQL architecture, query optimization, indexing strategies, replication, backup and recovery, security, monitoring, and performance tuning. The guide bridges the gap between development and database administration, helping developers write better queries and understand database behavior.

  15. 15
    Article
    Avatar of phoronixPhoronix·30w

    Valkey 9.0 Released With Ability To Achieve One Billion Requests / Second

    Valkey 9.0 has been released, marking a major milestone for the Linux Foundation-backed Redis fork. The new version introduces significant performance improvements, with the capability to handle up to one billion requests per second. This key-value database continues to evolve as a popular alternative in the data storage ecosystem.

  16. 16
    Article
    Avatar of stitcherstitcher.io·30w

    Stitcher.io

    A refactoring case study that simplified a content scheduling system by eliminating a state transition and cron job. The original design used three states (PENDING, SCHEDULED, PUBLISHED) with automated transitions, requiring cron jobs, console commands, and complex logic. The refactor removed the SCHEDULED state and added a future-dated publicationDate field to PUBLISHED posts, using a SQL query to find available time slots. This eliminated moving parts like cron jobs and automatic state transitions, trading one type of complexity for another. The key insight: modeling software directly from human processes doesn't always yield the simplest technical solution.

  17. 17
    Article
    Avatar of hashrocketHashrocket·29w

    PostgreSQL 18's UUIDv7: Faster and Secure Time-Ordered IDs

    PostgreSQL 18 introduces native support for UUIDv7, a new identifier format that combines the security benefits of random UUIDs with time-based ordering. Unlike UUIDv4's completely random values, UUIDv7 embeds timestamps in the first portion, enabling chronological sorting without separate created_at columns and improving index performance by reducing fragmentation. The format maintains collision resistance and security against enumeration attacks while delivering better database performance through sequential-like insertion patterns.

  18. 18
    Video
    Avatar of codinggopherThe Coding Gopher·31w

    99% of Developers Don't Get PostgreSQL

    Deep dive into PostgreSQL's internal architecture, covering its ACID implementation, object-relational model, and storage mechanisms. Explains multi-version concurrency control (MVCC) for handling concurrent transactions without blocking, the TOAST system for managing oversized data, and write-ahead logging (WAL) for durability and crash recovery. Includes details on heap files, page organization, tuple versioning, vacuum processes, and checkpoint mechanisms that ensure data consistency and performance at scale.

  19. 19
    Article
    Avatar of metalbearMetalBear·29w

    Introducing DB Branching in mirrord: Run Against a Shared Environment With a Personal, Isolated Database

    mirrord introduces DB Branching, a feature that creates temporary, isolated database branches for testing schema changes and migrations safely. When enabled, it automatically overrides database connection strings to point to a separate branch that mirrors the main database, allowing developers to test changes without affecting shared staging environments. The feature currently supports MySQL databases and is available in mirrord for Teams, with a step-by-step guide demonstrating how to test schema changes using a Go service on Kubernetes.

  20. 20
    Article
    Avatar of hnHacker News·33w

    Why TigerBeetle is the most interesting database in the world

    TigerBeetle is a financial transactions database built from scratch with modern distributed systems principles. It uses debits and credits as first-class primitives instead of SQL, achieving massive performance gains by packing 8,190 transactions into a single query. The database is written in Zig, distributed by default with Viewstamped Replication consensus, and handles storage faults through Protocol-Aware Recovery. TigerBeetle's development leveraged Deterministic Simulation Testing (DST) via their VOPR cluster running on 1,000 CPU cores, enabling them to build a Jepsen-validated system in just 3.5 years. Their engineering methodology, called TigerStyle, emphasizes assertions, static memory allocation, zero dependencies, and thinking deeply about performance during the design phase.

  21. 21
    Video
    Avatar of youtubeYouTube·30w

    Design Spotify Like a Senior Engineer (System Design Interview)

    A comprehensive walkthrough of designing a music streaming platform like Spotify, covering requirements gathering, capacity planning, high-level architecture, API design, database schema, and scalability strategies. The guide demonstrates how to handle 500,000 users and 30 million songs initially, then scale to 50 million users and 200 million songs using techniques like CDN distribution, database sharding, read replicas, and regional blob storage. Emphasizes the importance of planning and architecture over immediate coding, highlighting the skill progression from mid-level to senior engineering roles.

  22. 22
    Article
    Avatar of laraveldevLaravel Dev·32w

    Laravel Seeder & Migration Best Practices

    Eight practical guidelines for managing Laravel database migrations and seeders effectively. Covers reversible migrations, avoiding edits to deployed migrations, preventing duplicate seeded data using methods like updateOrCreate, organizing seeders into focused classes, separating development and production seed data with environment checks, leveraging factories for test data, and using migrate:fresh --seed for clean rebuilds. Emphasizes idempotent seeders and maintaining clean database history for stable, reproducible environments.

  23. 23
    Article
    Avatar of baeldungBaeldung·29w

    Introduction to BaseX

    BaseX is a lightweight XML database that stores, queries, and manipulates XML data using XQuery and XPath. It offers multiple interfaces including a GUI, command-line tool, and HTTP REST API. The tutorial covers installation requirements (Java 17+), basic database operations (creating, opening, dropping databases), resource management (adding, deleting XML files), querying with XQuery, and HTTP server setup with authentication. The REST API enables programmatic access to databases and resources, supporting CRUD operations through standard HTTP methods.

  24. 24
    Article
    Avatar of arangoArangoDB·32w

    ArangoDB: Multi-Model Database for Your Modern Apps

    ArangoDB is a multi-model database that combines graph, document, key-value, and search capabilities in a single system with a unified query language (AQL). It offers flexible deployment options including fully-managed cloud service (ArangoGraph), on-premises, and Kubernetes support across major cloud platforms. The database provides native client libraries for multiple programming languages and emphasizes database consolidation by replacing multiple specialized databases with one unified solution.

  25. 25
    Article
    Avatar of flydotioFly.io·33w

    Litestream v0.5.0 is Here

    Litestream v0.5.0 introduces major performance improvements and point-in-time recovery for SQLite databases. The update replaces the old WAL-based backup system with a new LTX file format that enables transaction-aware backups, hierarchical compaction, and faster restoration. Key changes include eliminating the generations concept for simpler backup management, supporting restoration from an average of just a dozen files, and adding NATS JetStream as a replica option. The upgrade is backwards compatible with existing configurations, though it can't restore from old v0.3.x WAL files. Future plans include a VFS for instant read replicas that can serve pages from S3 while hydrating in the background.

See all Database archives