Atatus APM delivers ready-to-use dashboards with real-time metrics, detailed transaction traces, slow database queries, and inefficient network calls. Quickly uncover root causes and resolve issues to enhance application performance and reliability.
Click a span to see what Atatus records.
Query text, call count and time, tied to the endpoint that ran it.
SELECT * FROM order_items WHERE order_id = $1
Click a deploy to compare the 15 minutes before and after it.
Click a deploy marker or use the arrow keys · shaded: 15 min before and after · sample data
| 15 min window | Before | After |
|---|---|---|
| Avg response time | 236 ms | 331 ms |
| Throughput | 2.3k rpm | 2.3k rpm |
| Failure rate | 0.21% | 0.65% |
| Error events | 18 | 160 |
142 more error events in the 15 minutes after this deploy.
POST /v2/projects/:projectId/deployments · revision=a91c3f2 · release_stage=productionOne screen from trace list to service map to JVM health.
Or just ask. The Atatus AI agent does all four from plain English, like “Why is payment-service slow since the last deploy?”
Requests, failures and Apdex for every day, week or month.
Click a day to read it alone · lowest Apdex in red · Apdex threshold T set in Project settings · sample data
Switch lens to see which endpoints burn the most time, and which ones have a long tail.
Average response time × throughput: the endpoints burning the most time overall.
$ npm install --save atatus-nodejs# require("atatus-nodejs") first, then start it with your license key# restart the app and send a few requests✓ agent connected · checkout-service · production✓ first trace received · POST /api/checkout · 1.24 s ├ order-svc buildOrder() 420 ms ├ postgres SELECT × 38 360 ms └ payment-svc charge() 760 ms
Pricing, deployment, data and the agent. If something is missing, a person will answer it on a demo call.
Atatus APM is priced per host, with unlimited applications and unlimited team members, so there is no per-user pricing. One agent per language instruments transactions, database queries, external calls and errors, and the same project gives you deployment tracking, Apdex and SLA reports. Atatus can also run on-premises when your data has to stay in your own network.
Yes. The Databases view shows query time and throughput for each database your application calls, along with its slowest queries. For a single transaction, Atatus splits the time into database, external and application spans with call counts and their share of the total, and each trace lists every database call with its query text. That is how a query that runs dozens of times in one request shows up.
The Atatus agents trace each request from entry to completion and follow it across your services, so a call from the API gateway to the order service, its database and a payment API shows up as one trace with timing for every span. You can see which service added the latency and which span failed. The same trace data builds the service map, where each instrumented service, database, queue, cache and external API appears.
Not if you already use OpenTelemetry. Atatus accepts traces, logs and metrics from OpenTelemetry instrumentation or the OpenTelemetry Collector, so you point the exporter at the Atatus endpoint with your API key. If you use a vendor agent, replace it with the Atatus agent for your language: install the package, set your license key and app name, and restart the application. Metrics start streaming in within seconds.
Yes. Atatus offers an on-premises setup for teams with strict compliance, data sovereignty or security requirements, so all monitoring data stays in your private network. You receive a Docker-based deployment package, install it, and point your Atatus agents at your own server URL. Contact the Atatus team to get access.
Atatus has APM agents for Node.js, Python, Java, .NET, PHP, Go and Ruby. Supported frameworks include Express, Koa, Hapi and NestJS for Node.js; Django, Flask and FastAPI for Python; Spring Boot, Tomcat and JBoss for Java; Laravel, Symfony and WordPress for PHP; Gin and Echo for Go; and Rails and Sinatra for Ruby. Calls to databases such as PostgreSQL, MySQL, MongoDB, Redis and Elasticsearch are captured too.