Dynatrace offers both SaaS and Managed deployments, but Grail and much of the current platform are SaaS-only. Kloudfuse runs inside your cloud and prices on one axis: telemetry ingested.
The biggest differences come down to how the platform is deployed, how usage is priced, and what happens as data volume and complexity grow.
Dynatrace Managed is still sold and has no announced end of life. Grail is not planned for Managed, so an on-premises deployment does not reach most of the capability Dynatrace has built on Grail. The choice is between running in your own data centre and running the current platform.
Metrics ingest at $0.15 per 100,000 datapoints, with no free custom-metric allowance on the rate card. Every extra dimension multiplies time series, which multiplies datapoints, which is a straight-line cost increase.
The bundled "retain with included queries" rate of $0.02 per GiB-day is available only for the first 10 to 35 days of retention. Past that you drop to $0.0007 per GiB-day plus $0.0035 per scanned GiB in query charges.
| Full-Stack host | $58 per 8 GiB host / month | A memory-optimised instance costs several multiples of a small one |
|---|---|---|
| Infrastructure host | $29 per 8 GiB host / month | |
| Foundation host | $7 per 8 GiB host / month | |
| Log ingest | $0.20 per GiB | |
| Metrics | $0.15 per 100,000 datapoints | No free custom-metric allowance; every added dimension multiplies datapoints |
| Traces and events | $0.20 per GiB | |
| RUM | $2.25 per 1,000 sessions | |
| Log retention | $0.02 per GiB-day | Bundled query rate, first 10–35 days only; after that $0.0007 per GiB-day plus $0.0035 per scanned GiB |
Dynatrace list pricing, August 2026. View current pricing →
Where the data plane runs, what each meter charges for, and what each platform will and will not do. Where a vendor does not publish something, the table says so.
| Dynatrace | Kloudfuse | |
|---|---|---|
| Where it runs | ||
| Data plane location | SaaS is the strategic model. Dynatrace Managed is still offered for mandatory on-premises requirements, and DPS covers hybrid — but Grail is not planned for Managed, so on-prem deployments cannot reach most of the current capability set. | Inside your VPC. Ingest, storage, query engine and the AI runtime all deploy into your own cloud account, writing to your object storage under your keys. Only metadata reaches Kloudfuse. |
| AWS, GCP and Azure | Yes. AWS · GCP · Azure | Yes. AWS · GCP · Azure |
| Drop-in migration from their agent | Not published. OneAgent | Yes. Datadog and OpenTelemetry agents accepted |
| What it costs | ||
| What you are billed on | Per host-hour, scaled by host memory, in three tiers — plus separate ingest meters on logs, metrics, traces, events and RUM, and separate retain and query meters on each of those. | One axis: terabytes ingested, across every signal. Storage and compute land on your own cloud bill because the platform runs in your account. |
| High cardinality | Priced, not capped. Every added dimension multiplies datapoints at $0.15 per 100,000 ingested, with no free custom-metric allowance. No published hard dimension limit per metric. | No cardinality premium, no active-series cap, no per-custom-metric charge. Adding a label does not change the bill. |
| Retention | 15 months of 1-minute metric granularity included; beyond 462 days you pay per GiB-day. Logs configurable from 1 day to 10 years, but the bundled query allowance only covers the first 10–35 days. | Your bucket, your lifecycle policy. Long retention is a storage decision rather than a pricing tier. |
| Your data | ||
| Sampling | Not published as a default rate. | No forced sampling. Full fidelity across metrics, logs, traces, events, RUM and profiles. |
| Every signal in one store | Partly. Grail unifies it — and Grail is SaaS only | Yes. One data lake |
| Log pattern reduction at ingest | Not published. Not published as an ingest control | Yes. Fingerprinting at ingest |
| Works with existing Grafana dashboards | Not published. No first-party datasource | Yes. Kfuse datasource |
| Portability | OneAgent and DQL. Deep proprietary instrumentation is what makes Davis good and also what makes leaving expensive. | Built on OpenTelemetry, with Grafana-compatible dashboards through the Kfuse datasource and support for TraceQL and PromQL. Your data stays in your object storage. Your instrumentation and dashboards remain portable and independent of Kloudfuse. |
| AI | ||
| Agents and assistants | Davis causal AI plus Dynatrace Intelligence and Intelligence Agents, launched January 2026. Worth noting that Davis is not a separate line item on the DPS rate card — the old Davis data unit model is legacy. | Dexter runs inside your boundary, on the model you choose, including a local one. Prompts and completions stay inside the boundary. A human stays in control. |
| Inference inside your own boundary | No. Davis runs in Dynatrace | Yes. Dexter runs in your VPC |
| Security and compliance | ||
| Government and regulated | FedRAMP Moderate since 2020, plus StateRAMP Moderate and IRAP. FedRAMP High was announced as an intent to pursue on 2 July 2026, not an authorisation. PCI DSS is mentioned on a marketing page but is absent from the trust centre list. | The data plane runs inside whatever boundary you have already had authorised. FIPS 140-3 validated cryptography (CMVP #5186, #5209); FedRAMP-targeted installations run with EBS encryption. |
| SOC 2 Type II | Yes. Type II | Yes. Type II |
| HIPAA | Not published. Not on their trust-centre list | Partly. In progress |
| FIPS-validated cryptography | Not published. Not published | Yes. 140-3 · CMVP #5186, #5209 |
| Data residency | Partly. Region choice, inside their boundary | Yes. Structural — your own VPC |
YesPartlyNoNot published
No. There is no announced end of support or end of life, and Dynatrace still positions it for customers with mandatory on-premises requirements. The real constraint is different and more consequential: Grail is not planned for Managed, so the on-prem product does not get the current platform.
Not as a line item. There is no Davis SKU on the Dynatrace Platform Subscription rate card — the Davis data unit model belongs to the classic pre-DPS licensing. You pay for Davis through the underlying data axes instead. Davis is included through the underlying Dynatrace data model rather than a separate AI SKU. Its causal analysis benefits from Dynatrace’s proprietary instrumentation; that same dependency can increase migration effort.
It is more honest in one direction and more expensive in another. A small instance costs less than it would under a flat per-host charge; a large memory-optimised instance costs several multiples. If you run big JVMs or in-memory databases, model it carefully rather than dividing your host count by anything.
No. Dynatrace holds FedRAMP Moderate, achieved in 2020. On 2 July 2026 it announced an intent to pursue High. Intent is not authorisation.
Dynatrace, Grail, Davis, DQL and OneAgent are trademarks of Dynatrace LLC. Kloudfuse is not affiliated with, endorsed by or sponsored by Dynatrace.
Run Kloudfuse against your own telemetry, in your own cloud, and see what a single correlated view does to your mean time to resolution.