AI Observability
Dynatrace is a unified observability and security platform that uses AI-powered automation to monitor, analyze, optimize, and secure applications, infrastructure, microservices, and user experiences across complex IT environments.
Publisher review
Dynatrace is a unified observability and security platform that uses AI-powered automation to monitor, analyze, optimize, and secure applications, infrastructure, microservices, and user experiences across complex IT environments. Founded in 2005 and headquartered in Waltham, United States, it serves over 15,000 customers in more than 30 countries, including enterprises managing distributed systems, microservices architectures, and hybrid cloud deployments. The platform targets IT operations, DevOps teams, and SRE professionals who need to manage cloud-native complexity at scale, offering full-stack monitoring, real user monitoring (RUM), synthetic testing, application security, and predictive analytics.
Its Davis AI Engine provides automated root cause analysis and anomaly detection across billions of dependencies, reducing alert noise and troubleshooting time. The Grail Data Lakehouse offers indexless, schema-on-read storage with massively parallel processing for observability, security, and business data at exabyte scale, enabling lightning-fast analytics using Dynatrace Query Language (DQL). PurePath Distributed Tracing provides code-level visibility into distributed transactions, tracking requests across services and microservices to identify performance bottlenecks.
Smartscape Topology Mapping creates real-time, dynamic visualizations of environment relationships and dependencies, automatically mapping interactions between applications, microservices, infrastructure, and cloud services. Dynatrace competes directly with Datadog, New Relic, and Splunk, positioning itself as the observability company for the AI era, transitioning from reactive analytics to predictive intelligence. According to its 2025 State of Observability Report, only 28% of organizations currently use AI-driven automation to align observability data with business outcomes, highlighting a market gap Dynatrace aims to fill.
However, the platform has notable trade-offs: costs can escalate quickly due to over-ingestion, the learning curve is steep, some features are missing or complex, and pricing is expensive compared to alternatives. For example, a Reddit user reported that for 500GB of logs daily, monthly cost comes to around $4,500 with 15 days of retention. Dynatrace uses a consumption-based pricing model with no penalties for exceeding commitments, offering multi-year and volume-based discounts. Its pricing tiers include Foundation & Discovery at $0.01 per hour per host ($7/month), Infrastructure Monitoring at $0.04 per hour per host ($29/month), Full-Stack Monitoring at $0.01 per memory-GiB-hour ($58/month per 8 GiB host), Container Observability at $0.002 per hour per pod ($1.40/month per pod), and Log Analytics at $0.20 per GiB ingest and process, $0.0007 per GiB-day retain, and $0.0035 per GiB scanned for queries.
How it works
-
AI-powered observability
Davis AI Engine provides automated root cause analysis and anomaly detection across billions of dependencies, correlating events across the entire stack.
-
Full-stack monitoring
Comprehensive observability for apps, microservices, and infrastructure includes APM, code-level profiling, and Kubernetes platform monitoring.
-
Application security
Runtime application security detects vulnerabilities and blocks threats like SQL injection, providing continuous security posture management.
-
Real user monitoring (RUM)
Tracks actual user interactions and session replays across web, mobile, and IoT applications to measure digital experience.
-
Synthetic testing
Simulates customer journeys and proactively monitors availability of applications and services from multiple geographic locations.
-
Predictive analytics
Uses AI to forecast potential issues before they impact users, enabling proactive remediation and reducing mean time to resolution (MTTR).
-
OpenTelemetry support
Ingests OpenTelemetry metrics and traces natively, allowing integration with existing instrumentation and multi-vendor environments.
Strengths and trade-offs
Strengths
- Ease of use: OneAgent auto-discovery instruments the entire stack without manual configuration, reducing setup time from days to minutes.
- Intelligent dashboards and AI-driven insights: Davis AI correlates billions of dependencies to pinpoint root causes, cutting alert noise by up to 90%.
- Comprehensive monitoring capabilities: Covers applications, infrastructure, containers, Kubernetes, logs, and user experience in a single platform.
- Effective debugging capabilities: Code-level tracing with PurePath and live code debugging via Code Monitoring enables instant troubleshooting in production.
Trade-offs
- Costs can explode fast: Over-ingesting data leads to massive bills; a Reddit user reported $4,500/month for 500GB of logs with 15-day retention.
- Steep learning curve: The breadth of features and DQL query language require significant training, especially for teams new to observability platforms.
- Complexity of features: Some advanced capabilities like Smartscape topology mapping and Grail data lakehouse are overkill for smaller environments.
- Missing features: Users report gaps in certain integrations and alerting customization compared to competitors like Datadog and Splunk.
Pricing context
Consumption-based model: Foundation $0.01/hour per host, Infrastructure $0.04/hour per host, Full-Stack $0.01/memory-GiB-hour, Container $0.002/hour per pod, Log Analytics $0.20/GiB ingest, $0.0007/GiB-day retain, $0.0035/GiB scanned. Multi-year and volume discounts available.
Getting started with AI Observability
-
Sign up for Dynatrace
Go to the Dynatrace website and create an account. Choose a pricing tier that matches your monitoring needs, such as Full-Stack Monitoring for comprehensive observability. Complete the registration process to receive your environment ID.
-
Install OneAgent on hosts
Download the OneAgent installer from your Dynatrace environment. Run the installer on each host you want to monitor, whether physical, virtual, or cloud-based. OneAgent automatically discovers and instruments applications, services, and infrastructure without manual configuration.
-
Connect data sources
Configure data ingestion by enabling log analytics, OpenTelemetry, or custom integrations. Set up log sources by specifying file paths or using the Log Monitoring API. For OpenTelemetry, configure exporters to send traces and metrics to your Dynatrace environment.
-
Configure Davis AI alerts
Define anomaly detection rules and alerting profiles in the Dynatrace settings. Use the Davis AI Engine to automatically correlate events and reduce noise. Set up notification channels like email, Slack, or PagerDuty to receive actionable alerts.
-
Create a dashboard
Navigate to the Dashboards section and create a new dashboard. Add tiles for key metrics such as response times, error rates, and infrastructure health. Use Dynatrace Query Language to filter and visualize data, then share the dashboard with your team.
Frequently Asked Questions
What is AI observability and how does Dynatrace use it?
AI observability uses artificial intelligence to monitor and analyze IT environments automatically. Dynatrace applies its Davis AI Engine for automated root cause analysis and anomaly detection across billions of dependencies, reducing alert noise and troubleshooting time for DevOps and SRE teams.
How much does Dynatrace cost per host for full-stack monitoring?
Dynatrace full-stack monitoring costs $0.01 per memory-GiB-hour, which is about $58 per month for an 8 GiB host. Infrastructure monitoring is $0.04 per hour per host ($29/month), and foundation monitoring starts at $0.01 per hour per host ($7/month).
What is the Dynatrace Davis AI Engine and what does it do?
The Davis AI Engine provides automated root cause analysis and anomaly detection across billions of dependencies. It correlates events across the entire stack to pinpoint issues, reducing alert noise by up to 90% and cutting mean time to resolution for IT operations and DevOps teams.
Does Dynatrace support OpenTelemetry for data ingestion?
Yes, Dynatrace natively ingests OpenTelemetry metrics and traces, allowing integration with existing instrumentation and multi-vendor environments. This enables teams to use open standards while leveraging Dynatrace's AI-powered observability and analytics capabilities.
What are the main weaknesses of Dynatrace for smaller teams?
Dynatrace costs can escalate quickly due to over-ingestion, with a reported $4,500 per month for 500GB of logs with 15-day retention. The platform has a steep learning curve, and advanced features like Smartscape topology mapping may be overkill for smaller environments.
How does Dynatrace compare to Datadog and New Relic for pricing?
Dynatrace uses a consumption-based model with no penalties for exceeding commitments, offering multi-year discounts. However, users report costs can be higher than alternatives, with log analytics at $0.20 per GiB ingested. Competitors like Datadog and New Relic may offer more flexible pricing for smaller deployments.
Alternatives
How AI Observability compares
Direct head-to-head against 3 competitors. Picked by 7wData.
AI Observability
- Pricing
- Consumption-based model: Foundation $0.01/hour per host, Infrastructure $0.04/hour per host, Full-Stack $0.01/memory-GiB-hour, Container $0.002/hour per pod, Log Analytics $0.20/GiB ingest, $0.0007/GiB-day retain, $0.0035/GiB scanned. Multi-year and volume discounts available.
- Target
- Dynatrace is a unified observability and security platform that uses AI-powered automation to monitor, analyze, optimize, and secure applications, infrastructure, microservices, and user experiences across
- Strength
- Ease of use: OneAgent auto-discovery instruments the entire stack without manual configuration, reducing setup time from days to minutes.
- Watch for
- Costs can explode fast: Over-ingesting data leads to massive bills; a Reddit user reported $4,500/month for 500GB of logs with 15-day retention.
Langfuse
- Pricing
- Free tier; Team $59/month; Self-hosted open source
- Target
- Teams wanting open-source, self-hosted LLM observability
- Deployment
- Self-hosted or cloud
- Strength
- Open-source with trace viewing, prompt versioning, and cost tracking
- Watch for
- Self-hosting requires DevOps effort; cloud tier limits data retention
LangSmith
- Pricing
- Free tier; Pay-as-you-go; Enterprise custom
- Target
- Teams heavily using LangChain for LLM development
- Deployment
- Cloud only
- Strength
- Deep LangChain integration for tracing and debugging chains
- Watch for
- Tied to LangChain ecosystem; limited standalone observability for non-LangChain apps
Fiddler
- Pricing
- Custom/Contact sales
- Target
- Enterprises needing ML compliance and model monitoring
- Deployment
- Cloud or on-premises
- Strength
- Hierarchical agent traces with real-time guardrails and compliance monitoring
- Watch for
- Pricing not public; complex setup for smaller teams
User reviews
No user reviews yet. Be the first to write one.
Sources
Reporting on this tool draws on these publicly available sources.