Elastic Observability 8.16: Enhanced OpenTelemetry support, advanced log analytics, and streamlined onboarding

Elastic Observability 8.16 announces several key capabilities:

• Amazon Bedrock integration for LLM observability  adds comprehensive monitoring capabilities for LLM applications built on Amazon Bedrock. This new integration provides out-of-the-box dashboards and detailed insights into model performance, usage patterns, and costs — enabling SREs and developers to effectively monitor and optimize their generative AI (GenAI) applications built on Amazon Bedrock in addition to existing support for applications that use Azure OpenAI.

• Unified Kubernetes observability with Elastic Distributions of OpenTelemetry (EDOT) delivers automated deployment and configuration of OpenTelemetry collectors through the OpenTelemetry Operator. This streamlined approach includes zero-code instrumentation options and preconfigured dashboards, allowing organizations to quickly gain comprehensive visibility into their Kubernetes environments without manual setup.

• Enhanced log analytics and streamlined onboarding introduce a context-aware Discover experience and new quickstart onboarding workflows. The improved Discover interface automatically adjusts data presentation based on content type, while the new onboarding workflows simplify the setup process for host monitoring, Kubernetes monitoring, and the Amazon Firehose delivery stream.

Elastic Observability 8.16 is available now on Elastic Cloud — the only hosted Elasticsearch offering to include all of the new features in this latest release. You can also download the Elastic Stack and our cloud orchestration products — Elastic Cloud Enterprise and Elastic Cloud for Kubernetes — for a self-managed experience.

What else is new in Elastic 8.16? Check out the 8.16 announcement post to learn more >>

The detailed logs view below provides full visibility into raw model interactions, capturing both the inputs (prompts) and the outputs (responses) generated by the models. This transparency enables you to analyze and optimize how your LLM handles different requests, allowing for more precise fine-tuning of both the prompt structure and the resulting model responses. By closely monitoring these interactions, you can refine prompt strategies and enhance the quality and reliability of model outputs.

The Amazon Bedrock overview dashboard provides a comprehensive view of the initial and final response times. It includes a percentage comparison graph that highlights the performance differences between these response stages, enabling you to quickly identify efficiency improvements or potential bottlenecks in your LLM interactions.

As with any Elastic integration, Amazon Bedrock logs and metrics are fully integrated into Elastic Observability, allowing you to leverage features like SLOs, alerting, custom dashboards, and detailed logs exploration.

We now use the OpenTelemetry Operator to automate the entire EDOT collector lifecycle, from deployment to scaling and updating. With automatic instrumentation via EDOT SDKs that support multiple languages like Node.js, Java, Python, and more, users can focus on applications instead of observability instrumentation.

This three-step flow simplifies the deployment of OpenTelemetry for Kubernetes with Helm. First, users add the OpenTelemetry repository to Helm for streamlined access. Then, the OpenTelemetry Operator is installed with a single command, automating the setup and configuration. Finally, optional instrumentation is made easy by auto-injecting libraries into annotated pods. This process enables fast, hassle-free observability for Kubernetes environments.

We have bundled all essential OTel components for Kubernetes observability, including receivers and processors. OTel-native Kibana dashboards give you comprehensive observability without manual configuration. By leveraging receivers like the Kubernetes and Kubeletstats Receivers, we now bring you turnkey observability that simplifies the monitoring process across Kubernetes environments.

EDOT Collector eliminates the need for an APM server, allowing trace data to flow directly into Elasticsearch via the Elasticsearch exporter. This reduces infrastructure overhead while maintaining rich, real-time performance insights. By consolidating APM functionality into the EDOT ecosystem, Elastic reduces operational complexity and costs — offering a streamlined, scalable observability solution.

This approach ensures we fully preserve OpenTelemetry’s semantic conventions and data structure, including resource attributes, for consistent and reliable observability.

The new summary column allows you to view important information at a glance. Service names are highlighted and important resource fields are displayed by default, followed by the log message, error, or stacktrace.

Similar improvements are also present for select fields, such as “log.level” and “service.name” when adding them individually as a dedicated column. The log level is highlighted based on the severity, and the service name also has the richer display state and offers direct links to the APM UI.

The Data Set Quality page has been extended with additional details to address common issues that result in _ignored fields. The UI now gives users additional details around the data quality issues. 

We're moving our hosts feature to general availability, helping you detect and resolve problems with your hosts more effectively. Key improvements include:

• Viewing hosts and their metrics detected by APM (even if you aren't explicitly observing them)

• Onboarding your hosts easily

• Gaining consistent metrics for your hosts in Observability, such as hosts, infrastructure inventory, dashboards

At the end of this guided workflow, users are provided a link to the appropriate prebuilt dashboard for each integration so that they can explore their data. 

Kubernetes monitoring: Sets up monitoring of the Kubernetes cluster and the container workloads using the standalone Elastic Agent. When a user follows this quickstart-guided workflow for Kubernetes monitoring from the Add Data page, they will obtain a Kubernetes manifest file with predefined defaults for logs and metrics collection. The System and Kubernetes integrations are also automatically installed in Kibana. Users can tweak the provided manifest file based on their needs and use their existing infrastructure-as-code tooling for the Agent lifecycle management in their production environment. 

At the end of this guided workflow, users are provided a link to the Kubernetes cluster overview dashboard so that they can explore the metrics and logs that have just been ingested.

Amazon Data Firehose (technical preview): This guided workflow simplifies the setup of an Amazon Data Firehose delivery stream using a prepopulated Amazon CloudFormation template, ingesting all of the available Amazon CloudWatch Logs and Metrics across multiple services for a given customer account.

Users can either use the AWS console or the AWS CLI to complete this guided workflow, as shown in the illustration below. Users are not required to provision or manage any agent as part of this workflow (agentless).

At the end of this guided workflow, users are provided a link to the appropriate prebuilt dashboard or curated UI to explore their data on a per-service basis. 

The host feature will help you by:

• Onboarding your hosts quickly with effortless onboarding journeys (via OTel)

• Seeing what needs attention using alerting and beginning RCA following alerting workflows
• Comparing host performance metrics to find the root cause
• Spotting dependencies by seeing which APM-instrumented services are running on your host
• Identifying resource bottlenecks by viewing the processes and threads (via Universal Profiling)

Our technical preview release of this capability will allow you to:

• View your hosts, containers, and services even if you only collect logs

• See what needs attention using alerting and beginning RCA following alerting workflows
• Perform seamless service analysis using workflows between Discover and Services

We're excited to announce that Elastic synthetic monitoring now includes first-class multifactor authentication (MFA) support, making secure testing of protected applications easier than ever. This enhancement empowers users to fully automate tests on secure applications without needing UI interactions for the generation of authentication codes — delivering smoother and more secure synthetic monitoring workflows on both inline- and project-based journeys. Learn more in our documentation.

Read about these capabilities and more in the release notes.

Existing Elastic Cloud customers can access many of these features directly from the Elastic Cloud console. Not taking advantage of Elastic on cloud? Start a free trial.

The release and timing of any features or functionality described in this post remain at Elastic's sole discretion. Any features or functionality not currently available may not be delivered on time or at all.

In this blog post, we may have used or referred to third party generative AI tools, which are owned and operated by their respective owners. Elastic does not have any control over the third party tools and we have no responsibility or liability for their content, operation or use, nor for any loss or damage that may arise from your use of such tools. Please exercise caution when using AI tools with personal, sensitive or confidential information. Any data you submit may be used for AI training or other purposes. There is no guarantee that information you provide will be kept secure or confidential. You should familiarize yourself with the privacy practices and terms of use of any generative AI tools prior to use. 

Elastic, Elasticsearch, ESRE, Elasticsearch Relevance Engine and associated marks are trademarks, logos or registered trademarks of Elasticsearch N.V. in the United States and other countries. All other company and product names are trademarks, logos or registered trademarks of their respective owners.