Apache Lucene is combining two optimizations for evaluating disjunctive queries, bringing significant speedups to top-k queries, especially with many terms and/or high-frequency terms.

Elasticsearch 8.9 introduces optimizations to cardinality aggregations by filtering out documents whose value has already been collected.

Data ingestion involves a lot of things that usually take a non-negligible amount of time. Fortunately, we've got improvements in 8.6, 8.7, and 8.8, which enabled some good speedups for end-to-end ingestion.

Elasticsearch 7.7では、以前のバージョンに比べてヒープ領域の使用量が大きく減少します。この進化はElasticsearchクラスターの安定性を高め、大きなサイズのデータを格納する際のコスト抑制に貢献します。

Elasticsearch 7.0はタイプを受け入れるAPIを廃止し、新たにタイプレスAPIを導入するほか、_default_マッピングをサポート対象外とします。