Dynatrace Pricing 2026: Per-Host Cost Guide

Dynatrace pricing 2026 per-host cost: Full-Stack Monitoring $58/mo per 8 GiB host ($0.01 per memory-GiB-hour); Infrastructure Monitoring $29/mo ($0.04/hr); Foundation & Discovery $7/mo ($0.01/hr); logs $0.20/GB; DPS pool commitment $25K+ annual minimum. Dynatrace bills on memory size, not host count, which is the single largest underestimated cost driver.

Last verified: Sep 16, 2026.

At a glance

• Foundation & Discovery: $7/host/mo ($0.01/host-hour)
• Infrastructure Monitoring: $29/host/mo ($0.04/host-hour)
• Full-Stack Monitoring: $58/8 GiB host/mo ($0.01/memory-GiB-hour)
• Application Security: $13/8 GiB host/mo
• Kubernetes pod: $1.40/pod/mo (free on Full-Stack hosts)
• Logs and traces: $0.20/GiB ingest

Side-by-side product pricing comparison

Dynatrace Platform Subscription (DPS) explained

Dynatrace does not sell products; it sells capability units drawn down by usage.

DPS is the licensing model where a customer commits to an annual minimum spend and consumes that commitment at published rate-card prices. Each product — Full-Stack Monitoring, Infrastructure Monitoring, Application Security, Log Management, Synthetic Monitoring, Business Analytics — draws down from the pool at a published rate (Dynatrace pricing page, August 2026).

The benefit of the pool model is flexibility: customers can flex consumption between products (more APM this quarter, more security next quarter) without renegotiating each line item. The drawback is that estimating consumption is harder than estimating per-host plus per-GB pricing. Unused units typically do not roll over.

DPS is the only published pricing model for Dynatrace in 2026. There is no longer a separate per-product SKU pricing. Customers on legacy contracts can convert, but the move to DPS is essentially mandatory for new enterprise procurement (Dynatrace pricing FAQ, October 2025).

Overages are billed at the same rates as pre-paid consumption, not penalty overages. Customers who consume more than their annual commit continue to use the platform on an on-demand basis billed monthly at the same rates. Customers can alternatively increase the commit spend to attain a higher discount tier.

Three product tiers: Foundation, Infrastructure, Full-Stack

The product tiers are differentiated by capability, not by underlying technology.

Foundation & Discovery at $7/host/mo ($0.01/host-hour): basic infrastructure monitoring and IT inventory. Includes basic host health indicators, host and process topology detection, filesystem monitoring, OS services monitoring, and collection of logs in context. Best for small-scale environments and teams starting their observability journey.

Infrastructure Monitoring at $29/host/mo ($0.04/host-hour): everything in Foundation plus process, disk, memory, and network analysis, plus custom metrics for extensibility. Best for production environments that need deeper visibility but do not require APM or distributed tracing.

Full-Stack Monitoring at $58/8 GiB host/mo ($0.01/memory-GiB-hour): everything in Infrastructure plus APM, automated root cause analysis for end-to-end transactions, code-level profiling, Kubernetes Platform Monitoring, OpenTelemetry metrics and traces, and 10 days of trace data retention extendable to 10 years. Best for enterprise observability with AI-driven root-cause (Dynatrace pricing rate card, 2026).

The memory-banded pricing of Full-Stack Monitoring is the most distinctive feature: a 16 GiB host costs $116/month and a 32 GiB host costs $233/month, double and quadruple the 8 GiB baseline. Teams that underestimate their host memory size are the most common source of bill shock in Dynatrace deployments.

Application Security, Kubernetes, and add-ons

Application Security and Kubernetes Platform Monitoring are typically add-ons to Foundation or Infrastructure tiers.

Application Security at $13/8 GiB host/mo ($0.0225/memory-GiB-hour): detects and remediates vulnerabilities in first-party and third-party code, automatic prioritization by risk and impact, ingest and analysis of external security findings. Application Security is the most common add-on for enterprises that pair observability with runtime security (Dynatrace pricing page, August 2026).

Kubernetes Platform Monitoring at $1.40/pod/mo ($0.002/pod-hour): Kubernetes metrics and events (CPU, memory, network, health views, topology). Included free on hosts running Full-Stack Monitoring. Best for Kubernetes-heavy environments without Full-Stack on every node.

Synthetic Monitoring and Business Analytics are billed as DPS capability units at tiered rates that vary by check frequency and complexity. Davis CoPilot (the AI root-cause engine) is included in Business Analytics capability units.

Enterprise Success and Support: priced as a percentage of annualized product fees with a $25,000 minimum. On large contracts, support alone can rival the cost of a smaller monitoring tool's entire bill.

Logs, traces, and RUM billing

Telemetry billing is the second-largest cost driver after Full-Stack Monitoring.

Logs and traces: $0.20 per GiB ingested at list. Stored in hot, warm, and cold tiers with separate query-time billing. Retention tier chosen at ingest time affects both query latency and storage cost.

RUM (Real User Monitoring): $0.00225 per session at list. Billed per browser or mobile session, not per user. High-traffic consumer-facing applications can drive RUM costs above observability costs.

Browser synthetics: $0.0045 per synthetic action. API synthetics: $0.001 per request. Ping synthetics are lower cost and used for uptime monitoring.

HTTP monitors: $0.001 per synthetic request. DNS and SSL monitors are similarly billed per check frequency.

The lesson: telemetry ingest and RUM sessions are the second-largest bill drivers after Full-Stack memory billing. For an enterprise at 1,000 hosts with Full-Stack and 100 GB/day logs, logs alone can be $600-$1,200/month at list (MonitoringCost Dynatrace analysis, 2026).

Real-world TCO: 100, 500, 1,000, and 2,000 hosts

Modular DPS means TCO depends entirely on which capabilities you turn on and the memory profile of your hosts.

100 hosts (Foundation only, 8 GiB average): 100 x $7 = $700/mo list. With typical 15-20% small-business discount: $560-$595/mo or $7K-$7K/yr.

500 hosts (Full-Stack, 16 GiB average): 500 x $116 = $58,000/mo list. With 25-30% mid-market discount on multi-year: $40K-$44K/mo or $480K-$525K/yr.

1,000 hosts (Full-Stack, 32 GiB average): 1,000 x $233 = $233,000/mo list. With 30-50% enterprise discount on 3-year commit: $116K-$163K/mo or $1.4M-$1.96M/yr.

2,000+ hosts (Full-Stack, 32-64 GiB average): $466K-$932K/mo list. Enterprise procurement above 2,000 hosts routinely negotiates 50-70% off list on extended commitments, bringing realistic spend to $200K-$400K/mo (MonitoringCost Dynatrace analysis, 2026).

Hourly pricing and 15-minute rounding

Dynatrace uses hourly metering with 15-minute rounding — the most cloud-native billing model in observability.

Dynatrace is the only major observability vendor to bill hourly with 15-minute minimum increments. A host running for 1 minute is billed for 15 minutes. A host running for 1 hour 3 minutes is billed for 1.25 hours. This rounding model is intended to be advantageous in cloud-native environments where hosts and services are rapidly spun up and destroyed (Dynatrace pricing FAQ, October 2025).

The hourly model also means that high-churn containerized environments can be billed more economically than with high-watermark per-host billing. Dynatrace customers running Kubernetes with high pod turnover typically see lower TCO than equivalent Datadog deployments because of the hourly metering.

The 15-minute rounding means very short-lived workloads (less than 15 minutes) are still billed for 15 minutes. This is a slight disadvantage for extreme autoscaling scenarios where pods live for less than 15 minutes.

How to choose between Dynatrace, Datadog, New Relic, and Grafana

The decision hinges on three variables: scale, memory profile, and AI-ops preference.

1. Scale. Below 100 hosts: Grafana Cloud Free/Pro or New Relic Standard. 100-500 hosts: Datadog or Dynatrace competitive; pricing depends on memory vs host count. 500-2,000 hosts: Dynatrace DPS competitive for memory-heavy enterprise workloads. Above 2,000 hosts: Datadog and Dynatrace both negotiable to similar effective rates.
2. Memory profile. Memory-heavy hosts (16-64 GiB average): Dynatrace DPS is expensive; Datadog per-host is cheaper. Memory-light hosts (<=8 GiB average): Dynatrace DPS is competitive. Containerized workloads with high turnover: Dynatrace hourly metering is favorable.
3. AI-ops preference. Want AI-driven root cause (Davis CoPilot) and OneAgent automated instrumentation: Dynatrace. Want flexible modular product mix: Datadog. Want consumption-based simplicity: New Relic. Want open-source-friendly stack: Grafana Cloud.

What enterprise buyers should do next

Three actions for organizations evaluating Dynatrace in 2026.

1. Audit your actual host memory profile. Dynatrace bills Full-Stack Monitoring on memory size, not host count. A 32 GiB host costs four times what an 8 GiB host costs. Get an accurate memory profile of your production fleet before negotiating.
2. Run a 3-year DPS commitment scenario. Multi-year DPS pool commitments (typically 3 years) above 500 hosts routinely discount 30-50% off published list. The savings on a 3-year commit can pay for the migration cost.
3. Compare Davis CoPilot root-cause ROI against Datadog Watchdog and New Relic AIOps. Dynatrace's AI root-cause engine is a real differentiator; evaluate whether the reduced MTTR on incidents justifies the premium over per-host observability tools.

What to watch next

Three near-term datapoints. First, the Dynatrace Q3 FY2027 earnings expected early November — commentary on consumption optimization and Davis CoPilot adoption will signal pricing trends. Second, the Application Security GA roadmap — Dynatrace has been expanding runtime vulnerability detection capabilities; new SKU pricing may surface in late 2026. Third, the Grafana Cloud competitive response — Grafana Cloud has been aggressively pricing its Application Observability product; Dynatrace will likely respond with bundled Full-Stack pricing.