When you’re navigating petabytes of observability data, looking for relevant information can become frustrating. That’s why we launched Dynatrace segments: a powerful yet simple way to focus on the data that’s important for your specific use cases.

In this blog post, you’ll find out how segments can be easily applied to help you work more efficiently, whether you’re troubleshooting problems, monitoring SLOs, or investigating security threats.

Work smarter in Dynatrace with segments

Think of segments as smart global filters. Segments are:

• Multidimensional: Combine and multi-select different data segments, such as region, team, service, or environment, and apply them instantly.
• Dynamic: Segments automatically adapt as your systems evolve, like when new services are added, new Kubernetes clusters are created, or team responsibilities change.
• Globally available: Selected segments persist as you move through Dynatrace Apps, so you can drill into data without reapplying filters.
• Secure and compliant by design: Segments are fully governed by existing access management policies, ensuring you only see the data they’re authorized to view. Any data outside those permissions is automatically filtered out, independently of which segment is selected.
• Built for scale: Segments are purpose-built for the demands of dynamic, cloud native environments and scale effortlessly to petabyte-level data volumes.

Segments promote self-service data access

Segments can be centrally managed and rolled out to provide a consistent, organization-wide view of data. At the same time, any user can create and share their own segments, whether within their own teams or across the organization.

Building segments doesn’t require complex logic, and upcoming ready-made segments will make getting started even easier. You can define segments around specific entities, such as services or hosts, to automatically include all relevant signals across data types. Alternatively, you can select specific data types individually and apply basic filtering.

However, segments also offer great flexibility for power users who want to create highly granular, dynamic filters based on attributes, tags, or specific field values, leveraging the full power of Dynatrace Query Language (DQL).

Applying segments is seamless and consistent across the platform. You can select segments directly from the top of any app or dashboard. For quicker access, the selector remembers your recently used segments, and you can pin frequently used segments to your menu. For more granularity, simply combine segments with other filters like the filter bar or dashboard variables.

Segments speed up troubleshooting in cloud native environments

Imagine you’re a developer responsible for multiple microservices or applications running across various regions, clusters, and environments. When something breaks or performance dips, you need to access the right data in the right context, quickly.

Segments help by narrowing the view to exactly what matters for you and your team, whether it’s a specific service, an environment, or a region. Instead of sorting through hundreds of unrelated logs or services, you get answers on the spot. This allows you to:

• Instantly understand the health of your services.
  In Services, filter by ownership to view only the services your team is responsible for. Immediately, you can assess key metrics like failure rate, response time, and throughput, then instantly drill down into related logs, traces, or infrastructure components such as Kubernetes clusters and namespaces. Segment context is preserved across all views and drilldowns, so there’s no need to reapply filters or recall complex query syntax.

  

• Triage and resolve problems more effectively.
  In Problems, segments allow you to isolate issues that affect your team’s services, ensuring that only relevant alerts are shown. Define segments based on environment or stage, such as production or staging, to pinpoint problems within a specific context. This makes it easier to compare behavior across environments and direct attention where it’s most needed.
• Break down SLOs to focus on what you own.
  Narrow the scope of your Service Level Objectives (SLOs) to reflect what your team is actually responsible for. With segments, you can filter SLO tiles on dashboards to only include the services, environments, or traffic patterns relevant to your team, such as staging tests or requests from a specific region. This ensures your SLOs accurately represent performance and reliability for your area of ownership. You can also configure alerts to trigger on segment-specific conditions, like an SLO breach in a particular region or service, helping teams take targeted action faster.
• Validate releases with Site Reliability Guardian (SRG) and Workflows.
  Once you’ve focused on the KPIs that matter to your team, use Site Reliability Guardian (SRG) to prevent regressions from reaching production. Segment-specific conditions can be defined directly within individual SRG objectives, enabling automated release validation aligned with your team’s performance standards, executed via Workflows. For example, you can restrict validation to logs from a specific service in a hardening environment, significantly reducing the volume of queried data while ensuring release quality without unnecessary noise.

  

Investigate vulnerabilities and security signals faster with segments

Because not all assets carry the same level of risk or compliance requirements, segments empower you to stay focused by isolating the systems that matter most. Grouping assets in segments opens the door to tiering strategies based on business criticality or risk level, such as production vs. test environments, or high-risk vs. low-impact systems.

This precision becomes even more powerful when combined with the Dynatrace Vulnerabilities app, which highlights those vulnerabilities that are actively exploited at runtime and present real threats, such as internet exposure or access to sensitive data.

Instead of being overwhelmed by hundreds of generic alerts, you significantly reduce the signal-to-noise ratio. For example, if you’re part of a financial services team, you may use segments to first zero in and focus on a critical, customer-facing credit card application service. With Davis® AI assessment, your team is alerted only to vulnerabilities that are exposed, exploited, and urgent. This allows you to respond quickly and to maximize your impact.

Segments also prove valuable during security incident investigations. If there’s a spike in failed login attempts, selecting a segment like app:customer-portal, env:prod, and region:us-east instantly narrows the scope. Within Notebooks or many other Dynatrace Apps, you can explore logs, traces, and metrics tied to that specific context, no manual filtering or tool-switching required. And because segments enforce access controls by design, sensitive data stays protected while investigations remain fast, focused, and compliant.

Boost your efficiency with enterprise-scale filtering

Segments make observability data instantly accessible and actionable for every team. Whether you’re analyzing data in Notebooks, tracking SLOs, troubleshooting in Dashboards, or automating with Service Reliability Guardian (SRG), segments keep you focused, reduce noise, and help you move faster.

Available on the latest Dynatrace, segments bring consistent, dynamic filtering across your entire observability workflow. For more on segments, have a look at our documentation or check out these resources to learn more.

• How segments empower central teams
• How to filter data effectively to find what matters
• How to create and use Segments in Kubernetes environments

The post Cut through the noise with segments: simple, powerful, and dynamic data filtering appeared first on Dynatrace news.

Empowering real-time insights through petabyte-scale data processing of all your data

Modern cloud native systems generate massive volumes of telemetry data, spanning logs, metrics, events, and traces. Extracting real-time, actionable insights and enabling meaningful alerting and root cause analysis requires more than just centralized data storage; it requires a scalable data pipeline. Without such a pipeline, even the most advanced analytics tools are constrained by data flow and preparation bottlenecks. To address these challenges, we launched OpenPipeline, our unified data ingest solution for the Dynatrace® platform. OpenPipeline is a customizable, scalable data pipeline for ingesting, transforming, and harmonizing data for reliable analysis and correlation.

With its latest enhancements, OpenPipeline supports real-time data processing even at petabyte scale, by reliably handling high volumes of logs, traces, and other telemetry. This level of scalable pipeline processing is essential for delivering timely insights, enabling Dynatrace to:

• Detect anomalies across distributed systems in real time
• Perform proactive security analytics and real-time vulnerability detection
• Offer real-time insights into your business processes

Scale is not only about the sheer amount of data signals. It’s also represented in the size of individual signals that can be handled. An increasing number of use cases, such as parsing JSON/XML request bodies, capturing detailed audit logs, exploring transactional data dumps, or analyzing input vectors from ML model logs, require the ability to efficiently process large log files. With the release of Dynatrace version 1.311, OpenPipeline now supports log records up to 10 MB in size, empowering you to use Dynatrace for high-volume use cases. For more information, have a look at this Dynatrace support for large log records blog post.

Break silos with unified ingestion across all data types

Alongside scale, OpenPipeline provides unified ingestion across all data types, removing silos, simplifying integration, and ensuring that all data, regardless of format or source, is enriched, contextualized, and processed in the same tool, ready for advanced analytics.

Recent enhancements to data type processing include:

• Spans: You can now configure how spans are processed, including dropping specific fields or entire records, and assign a security context for fine-grained, record-level access control. Spans can also extract metrics and route them into defined target buckets. Full ingest functionality for spans is coming soon.
• RUM data: Ingesting Real User Monitoring (RUM) data, including user events, sessions, and associated metrics, is currently in preview and will soon be made generally available for all Dynatrace SaaS customers.
• Events: OpenPipeline now supports custom processing rules for a broad range of event types, including security events, software development lifecycle (SDLC) events, business events, and Dynatrace internal system-level events.

Simplified pipeline management from setup to optimization

Historically, setting up pipelines for data ingestion involved time-consuming configuration of parsing, field mapping, and transformation logic for each data source. OpenPipeline now streamlines this process with ready-made processor bundles for widely used technologies and formats, accelerating data onboarding and standardizing data handling at scale for ingest sources such as:

• Hyperscaler support, including AWS and Azure
• Web servers like Apache, IIS, JBoss, HAProxy, or Nginx
• Programming languages like .NET, PHP, Java, Python, or NodeJS
• Databases and app frameworks like Elastic or Cassandra

With just a few clicks, you can apply a processor to a source, ensuring all fields are parsed correctly, attributes are renamed, and data structures are accurately standardized. Each bundle includes example records that can be tested interactively, with further customization available to meet specific requirements. This allows your teams to onboard new telemetry streams in minutes instead of hours. We’ll continue to expand our support to cover more technologies in the future. You can also create your own processors to handle any custom format.

Reveal hidden value with data transformation

Upon ingestion, modern observability, security, and business data can be structured in varying, often complex forms. OpenPipeline offers advanced transformation capabilities that allow for easy extraction of relevant data beyond the creation of simple events and metrics:

• Extract values from deeply nested JSON fields. For example, extract the threat identifier from a JSON-formatted security event and convert it into a security metric.
• Create unified metrics from different data types. For example, consolidate error codes from both logs and spans into a single metric for cross-service comparison.

Stay informed: Real-time visibility and alerting for your pipelines

To optimize pipelines, teams need visibility into their performance. OpenPipeline now exposes detailed metrics at key processing stages: ingest, routing, and output, as well as not-stored-records. With these metrics, you can instantly verify any pipeline configuration and detect anomalies early.

These metrics power:

• Real-time microcharts within the OpenPipeline user interface, showing data trends and ratios over the last 30 minutes.
• A ready-made dashboard used to explore daily or weekly summaries, historic volume trends, and detailed routing stats by pipeline or source.
• Smart alerting. Traffic fluctuations on ingest are common, so you can leverage AI-powered dynamic baselining to raise custom alerts and detect anomalies before an issue occurs.
• Automated operations that trigger notifications or perform autonomous remediation steps once an ingest anomaly or traffic volume deviation is detected.

Easy transition to OpenPipeline for existing customers

For Dynatrace SaaS customers using classic pipelines, we’ve simplified the transition to OpenPipeline. Start utilizing OpenPipeline for new data sources while keeping your existing log processing configurations fully operational and uninterrupted. This side-by-side, risk-free approach supports gradual adoption and allows you to modernize your data processing without disrupting ongoing workflows.

Get started today

OpenPipeline is now available to all customers running the latest version of Dynatrace SaaS. Already today, you can:

• Benefit from processing a broad range of data types, including logs, traces, events, and RUM data (currently in preview).
• Take advantage of processing the full payload of large log records, parse embedded XML or JSON structures, parse events or metrics from complex log records, or use them for deep search analytics and forensic use cases—all without splitting individual log lines into separate events. Leverage the built-in processor bundles to get started with popular formats.
• Explore the ready-made OpenPipeline dashboard on the Dynatrace Playground.
• Use real-time pipeline metrics to tune and optimize your configurations.

The post Latest OpenPipeline upgrades simplify high-volume, real-time data processing appeared first on Dynatrace news.

When you’re working with complex, multi-layered data, pinpointing the root cause of an issue or identifying anomalies can feel like trying to find a flickering light in a city skyline. Effective troubleshooting demands a deep understanding of data structures, filters, and query logic. But what happens when you’re staring at a mountain of logs or metrics and don’t even know what to look for? For instance, imagine you’re investigating a sudden spike in error rates across your application. Without knowing which service triggered the cascade or how to filter the noise, you’re left guessing where to begin.

With the introduction of deep search functionality in Grail, we’ve introduced a new way to cut through the noise of complex data. You can now explore data effortlessly without needing to understand every single layer of complexity in your data structure.

Investigating data is a time-consuming process

Difficult data investigations can be taxing. Examining data takes too long, and probing for issues is tricky for those who don’t know the complexities of the data structure. In the query example below, taken before the introduction of DQL search, we attempt to filter within spans where any event sent from an Oracle database contains the phrase “error”.

Search within complex structures

Grail’s unique data warping technology allows for index-free, schema-on-read, high-performance queries, enabling you to flexibly analyze diverse datasets. While Grail has long been capable of dynamically interpreting complex data structures at query time, with the addition of the DQL Search command, you can now deep dive into your metrics, logs, traces, and other data stored in Grail in a way more agile and intuitive way.

Simplified string-based filtering across multiple fields

You can now search across vast and complex data sets without needing to understand the underlying structure. By leveraging string-based search, the system offers intuitive exploration of deeply nested data—no manual filters or query syntax are required. This approach eliminates the traditional barriers of schema knowledge and filter construction, making data investigation accessible to everyone. (See Figure 2 below.)

Surface the needle in the haystack faster

You can search across all fields, not just indexed or predefined ones, making it easy to surface any data without needing to pre-tag or preprocess it. DQL search is optimized to be highly performant, even with terabytes of data.

Search in nested structures

Finding relevant data within nested elements or arrays is cumbersome, requiring a lot of effort to ensure you don’t exclude necessary parts of the data structure. With the new DQL Search command, you can search across complex data structures, including nested elements and arrays, to find exactly what you need.

Imagine a DevOps engineer troubleshooting a failed deployment in a development environment. Suspecting a misconfiguration or runtime error, the engineer needs to find traces containing the term “exception”—buried deep within the span.events array of nested records. The system scans across all nested fields and surfaces relevant spans instantly—saving time, reducing friction, and accelerating root cause analysis.

Now, consider another example with Dynatrace Security Investigator. You’re investigating a security incident in your Amazon cloud environment. You need to track a specific event from your AWS CloudFront logs, and you have the CloudTrail event ID value. You have no idea where to find the event ID. It could be buried in the content, tucked away in header values, or stored in an obscure field. The DQL Search command empowers you to search across all fields instantly—no filters, no field-mapping, just fast, accurate results when you need them the most.

As simple as using a search bar, inside DQL

Experience the flexibility and power of exploratory analytics in Dynatrace without worrying about indexes or schemas. The DQL Search command allows for deep, detailed investigations while maintaining simplicity, making it accessible to a wider range of users.

The post Deep Search: Transform your data exploration and investigation with Dynatrace Grail appeared first on Dynatrace news.