Elastic 8.16: BBQ, LLM observability, protect your cloud assets

The 8.16 release includes many high-impact capabilities across our portfolio:

• Elastic is the first vector database with Better Binary Quantization (BBQ) for large workloads. 

• Rock-solid large language model (LLM) observability for Amazon Bedrock comprehensively monitors and resolves LLM invocation errors and performance challenges — extending visibility into another popular LLM.

• Protect your cloud assets with a SIEM that finally solves cloud protection and contextual investigations in one license. 

• Explore an updated Kibana experience that adapts to context and is a snap to navigate.

Read about the highlights by solution below.

Elasticsearch helps developers build AI-powered search experiences on the industry's most used vector database with out-of-the-box semantic search and generative AI capabilities.

Highlights in 8.16:

• Elastic is the first to offer BBQ — Better Binary Quantization is a performant and efficient optimization for vector databases that is now in technical preview. It enables superior recall and customizable options for large workloads with scalar quantization and bit vector support.

• Production-ready hybrid conversational search with retrievers and reciprocal rank fusion (RRF) is now generally available (GA), enabling easy result normalization and composability.

• Get started with Elastic in minutes with start-local, a one-step local deployment for Elasticsearch and Kibana — now open source again! Then, scale quickly with the lightweight Open Web Crawler for data ingestion. 

• Elastic AI Assistant for Search enables search users to gain contextual help from generative AI models on building queries, ingesting data, and creating AI-driven search experiences using Elasticsearch.

Find more details in the Elasticsearch 8.16 highlights blog or in the 8.16 release notes.

Elastic Observability prevents outages with search-based relevance, no-compromise data retention, improved operational efficiency and cost, and an open and future-proofed investment. 

Highlights in 8.16: 

Monitoring LLM performance

• LLM observability has extended to Amazon Bedrock for comprehensive monitoring and resolution of LLM invocation errors and performance challenges (technical preview). 

OpenTelemetry (OTel) data ingestion that just works

• Application auto-instrumentation with Elastic Distributions of OpenTelemetry (EDOT) SDKs — commercially supported OTel SDKs — is now supported through the OTel Operator for Go, Python, .NET (technical preview), and Java (GA).

• Quickstart workflows for Kubernetes monitoring using the EDOT OTel Collector and the Elastic Agent are now available, as well as out-of-the-box OTel-driven Kubernetes dashboards (technical preview).

• OTel data ingestion that just works — enabling dashboards, log analytics, APM, and AI-driven insights all with ECS compatibility and no schema conversions.

• Amazon Firehose quickstart workflow with CloudWatch metrics and logs support is now in beta. 

Log analytics enhancements

• Discover gives you customized views of your data and a summary column for at-a-glance and contextual logs exploration (GA).

• Improved log data quality with a “fix-it” workflow addresses common log ingest issues that result in _ignored fields — offering a quick and easy way to get more value from your log data (beta).

Universal Profiling

• Running Universal Profiling on the self-managed Elastic Stack is now generally available. On-prem users can now install and run Universal Profiling on ECE, Kubernetes (via Helm charts), and self-hosted Elastic stack deployments. An Enterprise license is required to run Universal Profiling outside of Elastic Cloud.

Find more details in the Elastic 8.16 Observability highlights blog or in the 8.16 release notes.

Elastic Security future-proofs SecOps with AI-driven security analytics, accelerating threat detection, investigation, and response.

Highlights in 8.16: 

• Extended cloud security provider protections (GA): Leverage native cloud security workflows with integrations for Wiz, AWS Security Hub, and Falco — enabling out-of-the-box contextual threat detection and investigations.

• Custom knowledge integration (GA): Elastic AI Assistant for Security now supports custom knowledge sources — delivering tailored responses for a more personalized experience.

• Agentless cloud security posture management (CSPM) integration (beta): Eliminate the need to manage agents with agentless CSPM — streamlining data ingestion.

• Expanded session view support (GA): Session view now includes Auditbeat and Auditd Manager — empowering analysts with deeper insights for comprehensive Linux process investigations.

Find more details in the Elastic Security 8.16 highlights blog or in the 8.16 release notes.

The Elastic Search AI Platform combines the precision of search and the intelligence of AI. All users — regardless of use case — can benefit from core enhancements.

Highlights in 8.16: 

• Explore an updated Kibana experience that adapts to context, is a snap to navigate, and provides a more responsive dev console with an all-new interface.  

• Stay in command of your data with Elasticsearch Query Language (ES|QL) — recommended queries, named parameters, and faster performance. 

• Try flexible chunking strategies with word- or sequence-based strategy, customizable maximum sizes and overlap parameters, support for streaming APIs for our inference API integrations, and anomaly detection jobs that automatically adapt to daylight saving time. 

Find more details in the Elastic Search AI Platform highlights blog or in the 8.16 release notes.

Ready to get started? Elastic 8.16 is now available on Elastic Cloud — the hosted Elasticsearch service that includes all of the new features in this latest release.