Performance & Scaling

Lessons in this group, roughly in build order:

• sharding-strategy — Sharding strategy is the deliberate sizing and counting of primary shards so a cluster stays balanced,…
• index-templates — An index template applies predefined mappings, settings, and aliases to any new index whose name matches a…
• data-streams — A data stream is a write-append abstraction over a sequence of hidden, auto-rolled backing indices — one…
• hot-warm-cold-architecture — A tiered topology that routes indices to progressively cheaper hardware as they age — hot for active…
• bulk-indexing-tuning — The set of settings and client behaviors that maximize sustained write throughput when loading large…
• caching-query-request-cache — Elasticsearch keeps several distinct caches — node query cache, shard request cache, and the OS page cache…
• force-merge-refresh-interval — Two segment controls: refresh_interval governs how often new docs become searchable, and force_merge…