elasticsearch

This skill acts as an expert Elasticsearch Database Administrator (DBA) assistant, designed to help developers and SREs architect, optimize, and maintain high-performance search and analytics clusters. It provides deep technical guidance across the entire lifecycle of an Elasticsearch deployment, from initial schema definition to complex production troubleshooting.

• Expert design of indices, mapping templates, and field types (keyword vs text, doc values, dynamic mapping control) to prevent mapping explosion and ensure optimal storage efficiency.

• Comprehensive shard and replica strategy planning, including shard sizing (10-50GB targets), reindexing workflows, and alias-based migration strategies for zero-downtime changes.

• Performance tuning through query analysis, slow log diagnosis, aggregation optimization, and implementing search_after for efficient pagination.

• Production operations management including Index Lifecycle Management (ILM), hot-warm-cold architecture setup, snapshot/restore procedures, and proactive monitoring of heap, GC, CPU, and thread pool rejections.

• Troubleshooting and stability engineering, covering node-level performance bottlenecks, disk watermarks, cluster rebalancing, and capacity planning for evolving data volumes.

• Start by defining the cluster topology, data retention requirements, and specific latency SLAs to tailor the configuration to your use case.

• Always analyze search vs write patterns, as append-only log workloads require significantly different strategies than frequent update scenarios.

• Utilize standard tools like the Profile API, slow logs, and cluster health metrics as the primary evidence source for performance tuning requests.

• When planning migrations or upgrades, focus on alias-based cutovers to minimize impact and ensure backout plans are ready.

• Ensure you specify the Elasticsearch version, as syntax and index settings often vary between major releases (e.g., 7.x vs 8.x).