Category
Databases store and retrieve an application’s data, and the choice of engine shapes performance, consistency and cost. Relational databases like PostgreSQL and MySQL use tables, SQL and ACID transactions, while NoSQL stores such as MongoDB, Cassandra and Redis trade some guarantees for flexible schemas or horizontal scale. Indexing, replication and sharding are the main tools for keeping queries fast as data grows.
10 articles
Elasticsearch architecture from the inside: the inverted index and its term dictionary, segments and the near-real-time refresh, shards and replicas as first-class partitioning and replication, and…
Data partitioning explained: the umbrella concept under sharding, horizontal vs vertical partitioning, the partitioning vs sharding vocabulary sorted,…
Read/write separation: routing reads to replicas with the staleness audit, replica routing mechanics, replication lag as the governing…
Search engine system design: search vs database queries, when to add elasticsearch, the indexing pipeline that keeps the…
Object storage explained: the flat key-space and HTTP model, blob vs object naming, object vs block vs file…
SQL vs NoSQL decided on real constraints: what relational databases guarantee, the types of nosql databases, when to…
How do database indexes work: the b tree index, clustered vs non clustered indexes, composite index column order,…
Sharding vs replication compared: what each mechanism scales, when replication is enough, when to shard, and why production…
Database sharding explained: range, hash, directory, and geographic strategies, choosing a shard key, hot shards, cross-shard queries, and…
Database replication explained: leader-follower, multi-leader, and leaderless topologies, synchronous vs asynchronous trade-offs, replication lag, and the failover failure…
