elasticsearch/docs/reference/release-notes/highlights.asciidoc

[[release-highlights]]
== What's new in {minor-version}

coming::[{minor-version}]

Here are the highlights of what's new and improved in {es} {minor-version}!
ifeval::["{release-state}"!="unreleased"]
For detailed information about this release, see the <<es-release-notes>> and
<<breaking-changes>>.

// Add previous release to the list
Other versions:

{ref-bare}/8.14/release-highlights.html[8.14]
| {ref-bare}/8.13/release-highlights.html[8.13]
| {ref-bare}/8.12/release-highlights.html[8.12]
| {ref-bare}/8.11/release-highlights.html[8.11]
| {ref-bare}/8.10/release-highlights.html[8.10]
| {ref-bare}/8.9/release-highlights.html[8.9]
| {ref-bare}/8.8/release-highlights.html[8.8]
| {ref-bare}/8.7/release-highlights.html[8.7]
| {ref-bare}/8.6/release-highlights.html[8.6]
| {ref-bare}/8.5/release-highlights.html[8.5]
| {ref-bare}/8.4/release-highlights.html[8.4]
| {ref-bare}/8.3/release-highlights.html[8.3]
| {ref-bare}/8.2/release-highlights.html[8.2]
| {ref-bare}/8.1/release-highlights.html[8.1]
| {ref-bare}/8.0/release-highlights.html[8.0]

endif::[]

// tag::notable-highlights[]

[discrete]
[[stored_fields_are_compressed_with_zstandard_instead_of_lz4_deflate]]
=== Stored fields are now compressed with ZStandard instead of LZ4/DEFLATE
Stored fields are now compressed by splitting documents into blocks, which
are then compressed independently with ZStandard. `index.codec: default`
(default) uses blocks of at most 14kB or 128 documents compressed with level
0, while `index.codec: best_compression` uses blocks of at most 240kB or
2048 documents compressed at level 3. On most datasets that we tested
against, this yielded storage improvements in the order of 10%, slightly
faster indexing and similar retrieval latencies.

{es-pull}103374[#103374]

[discrete]
[[query_phase_knn_supports_query_vector_builder]]
=== Query phase KNN now supports query_vector_builder
It is now possible to pass `model_text` and `model_id` within a `knn` query
in the [query DSL](https://www.elastic.co/guide/en/elasticsearch/reference/current/query-dsl-knn-query.html) to convert a text query into a dense vector and run the
nearest neighbor query on it, instead of requiring the dense vector to be
directly passed (within the `query_vector` parameter). Similar to the
[top-level knn query](https://www.elastic.co/guide/en/elasticsearch/reference/current/knn-search.html) (executed in the DFS phase), it is possible to supply
a `query_vector_builder` object containing a `text_embedding` object with
`model_text` (the text query to be converted into a dense vector) and
`model_id` (the identifier of a deployed model responsible for transforming
the text query into a dense vector). Note that an embedding model with the
referenced `model_id` needs to be [deployed on a ML node](https://www.elastic.co/guide/en/machine-learning/current/ml-nlp-deploy-models.html).
in the cluster.

{es-pull}106068[#106068]

[discrete]
[[simd_neon_optimised_vector_distance_function_for_merging_int8_scalar_quantized_vectors_has_been_added]]
=== A SIMD (Neon) optimised vector distance function for merging int8 Scalar Quantized vectors has been added
An optimised int8 vector distance implementation for aarch64 has been added.
This implementation is currently only used during merging.
The vector distance implementation outperforms Lucene's Pamana Vector
implementation for binary comparisons by approx 5x (depending on the number
of dimensions). It does so by means of SIMD (Neon) intrinsics compiled into a
separate native library and link by Panama's FFI. Comparisons are performed on
off-heap mmap'ed vector data.
Macro benchmarks, SO_Dense_Vector with scalar quantization enabled, shows
significant improvements in merge times, approximately 3 times faster.

{es-pull}106133[#106133]

// end::notable-highlights[]