Key New Features in Elasticsearch 8.0
Elasticsearch 8.0 introduces major updates including 7.x REST API compatibility headers, default-enabled security with registration tokens, known issues on ARM/macOS, a preview KNN search API using dense_vector, storage reductions for keyword and text fields, faster geo indexing, PyTorch model support, and numerous other enhancements across aggregations, allocation, analysis, authentication, and core infrastructure.
Elasticsearch is a distributed, multi‑tenant search engine built on the Lucene library, offering an HTTP web interface and schema‑free JSON documents. It is written in Java and released under the SSPL+Elastic License, with official clients for Java, .NET, PHP, Python, Ruby, and many other languages.
New features in Elasticsearch 8.0 include:
7.x REST API compatibility : optional headers allow 7.x‑compatible requests and responses, easing migration to the new version.
Security enabled by default : security (authentication, authorization, TLS) is automatically configured on first start, and a registration token is generated for Kibana or new nodes.
Known issues : on ARM/macOS the initial elastic password and Kibana token are not generated; use bin/elasticsearch-reset-password -u elastic to set the password and bin/elasticsearch-create-enrollment-token -s kibana to create a token.
New KNN search API (preview) : the dense_vector field enables k‑nearest‑neighbor searches for recommendation and NLP relevance ranking, offering faster approximate searches on large datasets.
Storage savings : updated inverted index encoding reduces disk usage for keyword , match_only_text , and text fields (e.g., a 14.4% reduction for the message field in benchmarks).
Faster geo indexing : multi‑dimensional point indexing for geo_point , geo_shape , and range fields is 10‑15% faster.
PyTorch model support : users can upload externally trained PyTorch models for inference, bringing modern NLP capabilities to the Elastic Stack.
Other notable changes span many components:
Aggregations : removal of adjacency matrix, MovingAverage pipeline, deprecated time/term sorting, and date histogram interval.
Allocation : removal of include_relocations setting.
Analysis : cleanup of versioned deprecations and removal of delimited_payload_filter .
Authentication : always add file and native realms unless explicitly disabled; default cluster.routing.allocation.enforce_default_tier_preference set to true.
Cluster Coordination : removal of connection timeout and deprecated delayed state recovery support.
Engine : stricter handling of only_expunge_deletes and max_num_segments requests.
Features / CAT APIs : deprecated local parameter removed from _cat/indices and _cat/shards .
Features / ILM+SLM : default tier‑preference enforcement enabled.
Features / Indices APIs : default prefer_v2_templates set to true; removal of deprecated _upgrade API and include_type_name parameter.
Infra/Core : various path, bootstrap, Joda, and date‑format clean‑ups.
Packaging : removal of SysV init support, JAVA_HOME support, and requirement for Java 17.
For a complete list of changes and detailed documentation, see the official blog post: https://www.elastic.co/cn/blog/whats-new-elastic-8-0-0 .
Top Architect
Top Architect focuses on sharing practical architecture knowledge, covering enterprise, system, website, large‑scale distributed, and high‑availability architectures, plus architecture adjustments using internet technologies. We welcome idea‑driven, sharing‑oriented architects to exchange and learn together.
How this landed with the community
Was this worth your time?
0 Comments
Thoughtful readers leave field notes, pushback, and hard-won operational detail here.