Datadog DASH 2024 Insights: New Features for Observability, Security, and Performance

(YIUCHEUNG/Shutterstock)

At this year’s Datadog annual conference, DASH 2024, the company made a string of key announcements, including new platforms and product updates designed to enable teams to observe their environment, secure the workloads and infrastructure, and fix issues before they affect customers.  The conference took place on June 25–26th, 2024, at the North Javits Center in New York City. 

One of the key announcements is the general availability of the Datadog LLM Observability platform, which is designed to allow AI application developers and machine learning (ML) engineers to monitor, improve, and secure LLM applications. 

LLM agents have helped organizations to effectively adapt GenAI for their specific use cases. While the LLM applications can be a powerful tool in unlocking new product opportunities, there remains a pressing need for granular visibility into their behavior. Running these complex LLM workflows in production and at enterprise scale can make it challenging for organizations to evaluate model performance, manage security, and diagnose errors. 

Datadog logo

The newly launched Datadog LLM Observability platform offers a solution to these challenges. It can trace LLM application workflows from end to end to help users monitor, troubleshoot, improve, and secure LLM applications. The platform users can also monitor LLM application operation performance, evaluate LLM application’s functional quality, and security exposures. 

Datadog has also integrated a fully configurable OpenTelemetry (OTel) Collector and Datadog Agent, creating a unified observability solution. Teams can now access Datadog’s comprehensive monitoring solutions alongside OTel’s capabilities for comprehensive monitoring, faster onboarding, simplified management, enhanced security, and reliability. 

The cloud observability platform also introduced Log Workspaces, a workbench for extracting insight from log data. Building on the capabilities of Datadog Log Explorer, the new feature helps teams swiftly navigate large volumes of varied log data. Currently available in private beta, Log Explore enables users to compose complex queries visually, build and share powerful reports, and transform data for efficient remediation. 

Datadog users now also have access to Datadog Live Debugging, currently in beta, which enhances production bug fixing by offering contextual insights to pinpoint root causes swiftly. In addition, users can now make better data-driven UX design decisions with Product Analytics, which offers a variety of features including Heatmaps, Session Replay, and Sankey. 

The security-related announcements by Datadog included the general availability of Agentless Scanning, enabling teams to quickly detect and remediate security vulnerabilities across their cloud infrastructure without the need for agent deployment. Datadog now offers one-click remediation for misconfigured cloud resources through its Cloud Security Management platform. 

Now available in private beta, Datadog’s Data Security has been launched to help users discover sensitive data in their cloud data stores. This includes personally identifiable information (PII) in the cloud. The platform automatically highlights sensitive data in AWS S3 buckets and RDS instances, allowing users to fix security issues impacting these cloud resources.  

(La1n/Shutterstock)

New Datadog features and updates aimed at enabling users to take proactive actions for enhanced efficiency and performance were also announced at the event. The Datadog Kubernetes Autoscaling solves several changes in resource management and cost optimization. Available in private beta, the new Kubernetes Autoscaling feature automatically adjusts Kubernetes workloads to optimize performance and cost-efficiency based on real-time telemetry data. 

DevOps, SRE, Security, and IT Ops teams often face challenges in the form of overwhelming alerts, and needing to manage multiple tools and resources effectively to resolve high-stakes issues. This can lead to inefficiency and burnout. Datadog has unveiled On-Call to unify observability, paging, and incident response onto one platform. It integrates observability data with service and team ownership details, facilitating quick alert triage. 

Providing visibility into changes that could impact system performance or stability, the new Change Tracking feature streamlines incident response by highlighting relevant changes and potential remediation steps. This new feature is currently in private beta. It can track deployment, Kubernetes pod crashes, traffic anomalies, schema changes, faulty deploys, and feature-flag changes. 

With the suite of innovations announced at DASH 2024, Datadog continued to pioneer solutions that can streamline enterprise operations, fortify security, and build digital resilience. These new features and updates by Datadog could be crucial for enterprises to thrive in the complex digital landscape. 

Related Items 

Datadog Launches IT Event Management to Enhance AIOps Capabilities

Foundational Secures $8M to Bring AI Agents into Data Engineering

Datadog Expands Strategic Partnership with Google Cloud and Integrates with Vertex AI