# Last9 vs Dynatrace: A Dynatrace Alternative on Open Standards > Last9 vs Dynatrace compared on AI and MCP, what the AI costs and where it runs, DPS pricing, custom metrics, DQL and PromQL, OpenTelemetry, and mainframe and on-prem coverage, with where each one fits. Source: https://last9.io/compare/dynatrace/ Both have an AI assistant, agents, and an MCP server. Last9 uses your model provider, charges nothing per investigation, runs in your own cloud account, and bills one licence. Dynatrace reaches further into the enterprise estate, down to the mainframe, and draws everything from one annual commitment. - **Gartner Cool Vendor 2025:** Named a Gartner® Cool Vendor in AI for SRE and Observability, cited for its unified telemetry platform. - **Your model, your cloud:** The Last9 agent runs on your model provider, with no per-investigation charge, in your own cloud account. - **One licence:** No annual drawdown across meters. The infrastructure runs on your own cloud bill. ## Last9 vs Dynatrace at a glance The questions engineering teams ask when they compare the two. | | **Last9** | **Dynatrace** | |---|---|---| | Does it have an MCP server for coding agents? | ✓ Yes, a hosted MCP server since March 2025, for any MCP client. An agent can also add drop rules and edit dashboards through it | ✓ Yes, a hosted MCP server for DQL queries, problems, and forecasts. Config changes go through the dtctl CLI | | What does the AI cost? | ✓ No per-query or per-investigation charge | No AI licence, but every query the AI runs draws on your DPS commitment | | Can the AI use our model provider, in our own cloud? | ✓ Yes. The Last9 agent runs on your model provider or inference endpoint, and Last9 runs in your own cloud account | No. Dynatrace Intelligence runs on SaaS only, with no documented option to bring your own model | | How is it priced? | ✓ One licence. The infrastructure runs on your own cloud bill, at your negotiated rates | An annual DPS commitment, drawn down by host hours, log GiB ingested, retained, and scanned, metric data points, and more | | How are custom metrics billed? | ✓ Kept, with no sampling. 20M series per metric per day by default, and no charge per label or metric name | Per data point ingested, so splitting a metric by a dimension multiplies its data points | | Can I filter telemetry before I pay to store it? | ✓ Yes. The Control Plane drops, aggregates, and routes at ingest | ✓ Yes. OpenPipeline filters, transforms, and routes data before it lands in Grail | | Do I have to learn DQL, or can I use PromQL? | ✓ PromQL and LogQL, or none: build log queries in the query builder, or ask in plain English | DQL only. No PromQL | | Does it support OpenTelemetry? | ✓ OpenTelemetry-native. Send OTLP from any SDK or Collector | Ingests OTLP. OneAgent does not send its data to other backends | | Does it cover mainframes and on-prem data centres? | No mainframe tracing. Last9 runs in your own cloud account | ✓ Yes. OneAgent traces CICS and IMS on z/OS, and Managed runs in your own data centre | ## AI on your model, in your cloud - **Last9:** One assistant in the app, in Slack, and in your IDE through MCP. It investigates, builds a diagnosis dashboard, and hands you the link. - **Last9:** The production agent, Last9, responds to pages with context from Slack, Linear, and PagerDuty. It runs on your model provider or inference endpoint and picks the model for each step. Customer data is never used for training. - **Last9:** An agent can do more than read. Through MCP it can add a drop rule and stop ingesting the noisy source it found. - **Dynatrace:** Dynatrace Intelligence pairs Davis causal AI with Dynatrace Assist, SRE agents, and a hosted MCP server. The AI has no separate licence, but its queries draw on DPS, and it runs on SaaS only. - **Tradeoff:** Davis AI's causal root cause analysis and Smartscape's automatic topology have years of enterprise use behind them. Compare both on your own incidents. ## One licence instead of a drawdown - **Last9:** One licence. The infrastructure runs on your own cloud bill, at your negotiated rates. - **Last9:** Labels like customer_id or tenant stay on the metric. You pay for volume, not for metric names or label values. - **Last9:** Drop, remap, redact, forward, and aggregate metrics, logs, and traces at ingest, from the UI. Every rule is previewed against live data before you save it. - **Dynatrace:** DPS is an annual commitment drawn down on a rate card: host memory hours, log GiB ingested, retained, and scanned, metric data points, and more. OpenPipeline filters data before it lands in Grail. ## Open query languages, portable instrumentation - **Last9:** OpenTelemetry-native. Send OTLP from any SDK or Collector, and your instrumentation stays portable. - **Last9:** PromQL and LogQL natively, with a builder and plain-English Ask Mode for anyone who does not write queries. - **Dynatrace:** Ingests OTLP, but DQL is the only query language, and OneAgent data does not leave Dynatrace for another backend. - **Matters most for:** Teams on Prometheus and Grafana, and teams that want instrumentation they can take to any backend. ## Which one fits your team? **Choose Last9 if:** - You want AI on your own model provider, with no charge per investigation - You want one licence instead of an annual commitment drawn down across meters - You work in PromQL, LogQL, and OpenTelemetry, and do not want to learn DQL - Production data must stay in your own cloud account, managed for you - Your cloud-native teams run Kubernetes on OpenTelemetry, next to a Dynatrace estate **Choose Dynatrace if:** - You run IBM Z and need traces that reach CICS and IMS - Part of your estate runs in your own data centres, outside the public cloud - Smartscape topology and Davis causal AI are central to how you run incidents - You want one agent to instrument everything, without app teams changing code ## Moving off Dynatrace What the switch involves, and what Last9 does for you. - **Forward deployed engineers:** A Last9 engineer runs the move with your team. - **Instrument with OpenTelemetry:** OneAgent data does not forward to other backends, so services move to OpenTelemetry. Anything already sending OTLP or exposing Prometheus metrics needs only a new endpoint. - **Move what you rely on:** A Last9 engineer moves your dashboards and alerts with your team, and rewrites DQL queries in PromQL or LogQL. - **Start with the cloud-native estate:** Move your Kubernetes and OpenTelemetry services to Last9 first, and keep Dynatrace on the mainframe and on-prem systems for as long as you need. Compare the two on your own incidents as you go. ## The objections we hear most Each one is a fair reason to hesitate. ### Dynatrace already has Davis AI, agents, and an MCP server. Why switch for AI? AI alone is not the difference. What differs is the model, the cost, and where it runs: the Last9 agent runs on your model provider, has no per-investigation charge, and runs inside your own cloud account. ### We rely on Smartscape and automatic discovery. Last9 discovers services and maps dependencies from OpenTelemetry traces. Smartscape builds its topology from OneAgent across hosts, processes, and services, and has years of enterprise use behind it. If you use that depth every day, test it before you switch. ### Part of our estate runs on the mainframe or on-prem. Then Dynatrace covers more of it. OneAgent traces CICS and IMS on z/OS, and Managed runs in your own data centre, without Grail, DQL, or the agentic AI. Last9 runs in your own cloud account, not in a data centre, so you can run both: Last9 for the cloud-native services, Dynatrace for the rest. ### Is Last9 proven at our scale? Last9 regularly handles 500M+ events a minute and 60M+ concurrent users, and has monitored 12 of the world's 20 largest streaming events. It is SOC 2 Type II certified and PCI ready, with SSO and access controls. Teams across fintech, healthcare, gaming, media, and commerce run it in production, including Replit, Pine Labs, Tata 1mg, and Housing.com. ## FAQ ### Is Last9 a good Dynatrace alternative? Yes, for teams that want AI on their own model provider with no per-investigation charge, one licence instead of a DPS commitment, PromQL and LogQL, or data inside their own cloud account. Dynatrace fits better if you need mainframe tracing, on-premises deployment, or Smartscape's automatic topology. ### How does Last9's MCP server compare with the Dynatrace MCP server? The Dynatrace MCP server is hosted. It generates and runs DQL, investigates problems, and forecasts, and Dynatrace changes configuration through its dtctl CLI. Last9's hosted MCP server, available since March 2025, works with any MCP client, such as Claude Code or Cursor, and can add drop rules and edit dashboards, so an agent can change what is ingested. ### How does Last9's AI compare with Dynatrace Davis AI? Dynatrace Intelligence runs on SaaS only, and its queries draw on your DPS commitment. Both offer an assistant, an MCP server, and agents that investigate problems. The Last9 agent runs on your model provider or inference endpoint, has no per-investigation charge, and runs in your own cloud account. Davis AI's causal root cause analysis has the longer track record. ### How does Last9 billing compare with Dynatrace pricing? Dynatrace pricing is an annual DPS commitment, drawn down on a published rate card: host memory hours for Full-Stack, GiB of logs ingested, retained, and scanned, metric data points, and more. Last9 is one licence, and the infrastructure runs on your own cloud bill at your negotiated rates. ### How does Last9 compare with Dynatrace custom metrics? Under DPS, Dynatrace custom metrics are billed per data point ingested. Dynatrace does not bill per dimension, but splitting a metric by a dimension multiplies its data points, and older classic licences bill in Davis Data Units instead. Last9 keeps high-cardinality labels with no sampling, allows 20M series per metric per day by default, and has no charge per label or metric name. ### Does Last9 use PromQL instead of Dynatrace DQL? Yes. Dynatrace queries everything in DQL, with no PromQL, and ingests OpenTelemetry over OTLP. Last9 is OpenTelemetry-native and supports PromQL and LogQL natively, so queries and instrumentation stay portable. ### Is there a Dynatrace Managed alternative that runs in our cloud account? Yes. Last9 runs single-tenant in your own cloud account, in any region, and manages it for you. It does not run in your own data centre. Dynatrace Managed does, with Dynatrace Mission Control updating it remotely, but Managed has no Grail, DQL, OpenPipeline, or agentic AI. ### Can Last9 replace Dynatrace OneAgent and APM? For APM, yes. Last9 covers service discovery, dependency maps, database and external calls, Apdex, exceptions, and trace correlation, from OpenTelemetry. It has no single-agent equivalent of OneAgent and no mainframe tracing. ### Can Last9 run alongside Dynatrace? Yes. Move your Kubernetes and OpenTelemetry services to Last9 first, and keep Dynatrace on the mainframe and on-prem systems. The two run side by side, and a Last9 engineer plans the split with your team. ### How do I migrate from Dynatrace to Last9? Instrument services with OpenTelemetry, because OneAgent data does not forward to other backends. Anything already sending OTLP or Prometheus metrics needs only a new endpoint. A Last9 engineer runs the move with your team, including your dashboards and alerts.