Harness the power of Hosted Graphite to gain real-time insights into your Heroku applications and Postgres Databases.
Keep a constant watch on your Heroku environment with instant access to detailed metrics. Detect anomalies, track trends, and ensure your applications and databases are running smoothly.
Automated dashboards directly from your Heroku Log stream for your Dynos. Augment these with custom metrics to track your applications and effortlessly customize and visualize your data. Easily create graphs, charts, and reports tailored to your unique monitoring needs.
As your Heroku applications grow, Hosted Graphite scales with you. From small projects to large-scale deployments, our platform adapts to your monitoring demands.
Integrate seamlessly with your Heroku setup. Effortlessly connect your applications and databases to Hosted Graphite and start monitoring without any hassle. Hosted Graphite will never compromise the speed of your app. Send as much as you want without worrying about slowing down your users.
Elevate your monitoring game with Hosted Graphite! Sign up now for a free trial or provision our Heroku add-on and experience the future of Heroku monitoring today!
Learn More→
Because our system is your system.
MetricFire is a managed observability platform that helps teams monitor production systems with clean dashboards and actionable alerts. Delivering signal, not noise. Without the operational burden of self-hosting.
We host Graphite so you don’t have to! We handle and manage your metric collection, storage, alerting, and visualization services so you can focus on more important things.
Hosted Graphite provides ways for you to send data from your applications, servers, and infrastructure to plot that data on professional and easy-to-read custom dashboards.
We give you the tools you need to send large amounts of data and get accurate information back out, analyze it, alert on it, and share it. This alongside integrations with other popular services, account/team management options and offer great support.
You could self-host your monitoring and spend a few hours setting it up, send some test data, and feel good when the graphs start working. That’s the fun part. The less fun part comes later when scaling, version updates, or performance issues pull you away from real work and turn self-monitoring into a constant distraction. So we handle all of that for you and let your team focus on what actually matters.
We define a metric as a unique time-series metric namespace that is stored with our service. Sending updated counts for existing metrics in your account will not count as a new metric we bill on. We charge on the total number of unique metrics in an account, not the amount of data points the metric receives. Also, we do not bill separately between host (CPU) metrics, custom metrics, or tagged metrics - 1 metric is 1 metric.
This is the number of individual data points that are sent to each unique metric name space, per second. If you send a VERY high rate of DPPS, rate limiting or 'throttling' will be applied to protect the health of our systems.
If you have very heavy DPPS traffic, we can help you configure a pre-aggregation service for your metrics, like StatsD or carbon-c-relay.
When you sign up for a free trial, you get a full test of Hosted Graphite for 14 days including 500 total metrics, 5 user seats, alerts, dashboards, Hosted StatsD, logging options, and access to all integrations.
If you need help setting up a custom trial account for a PoC, please contact support@metricfire.com or book a call with us and we can help set this up for you.
Yes! You can share your dashboards with teammates, customers, or whoever you need. We don’t charge extra for this.
It takes about 60s for your metric data to be available for visualisation once it has been ingested and stored in our backend.
Yes, MetricFire supports Single Sign On (SSO) via SAML-enabled identity providers. Just reach out to us and we will enable this feature in your account, for free.
We understand there needs to be flexibility here. If you are consistently going over your metric limit, the data points that you are sending will be dropped.
We are happy to work with you as your monitoring partner so we’ll send you a message to offer support with managing your metrics, or help you find the plan level that suits you best.
Also, we have a dashboard and alerts automatically created in your account to let you know if you're getting close to your metric limit and if data points are being dropped.
Yes! We manage a Heroku Add-On and you can get dyno, router, postgres, redis, kafka, and process metrics with the HG Add-On, or with a dedicated Hosted Graphite account.
We provide email and live chat support for all plans and we answer quickly! Our team is made of experienced engineers committed to great service.
We are based in the USA but we have team members based in multiple countries so we can provide support during a wide range of hours. We provide support via email, live chat and technical support meetings.
We aggregate and store data at the following resolutions:
When querying data in your Dashboards, we map the following time spans to the following resolutions:
Yes, we support graphite tagging, have our own Tag API, and have a Tag Search UI feature so you can easily locate and manage your tagged metrics. Refer to our Graphite Tag doc for more details.
Yes! Our composite graphite alerting lets you combine multiple alert conditions (using AND / OR logic), so alerts trigger only when the full rule is satisfied.
Data is stored in ISO:27001 and SOC2 Certified Data centres located in Frankfurt, Germany. Metrics sent to us directly are stored on bare-metal database servers, provided by Hetzner. Learn more here.
We operate a distributed eventually-consistent database built upon open-source Riak. In case of failure, the workload is distributed and we store three redundant copies of all data at all aggregation resolutions. Queries will get back to the user with answers only when two different systems agree, which is also useful for fault tolerance. We also have backup, failover, and disaster recovery plans in place you never have to worry about data loss.