Sysdig announced today it has added Prometheus compatibility to the Sysdig Secure DevOps Platform through which dashboards based on data collected by the open source platform can be generated and continuously updated on the Sysdig monitoring service.
Payal Chakravarty, vice president of product management for Sysdig, says that while Prometheus is gaining traction among developers who rely on it to monitor applications while they are being developed, most IT operation teams want to be able to consume Prometheus data alongside data being generated by the rest of their application portfolio. As such, it makes sense to add support for PromQL, the query language for Prometheus, to enable IT operations teams to create dashboards, alerts and metric analytics from the data collected by Prometheus, he says.
That approach provides the added benefit of allowing organizations to preserve investments in Prometheus exporters, configurations, alerts and dashboards using open source Grafana analytics tools created during the application development process, adds Chakravarty.
In addition, Sysdig launched PromCat.io, a free repository of curated Prometheus exporters, dashboards and alerts. Sysdig offers documentation and suggested configurations for all PromCat integrations, including instances for Kubernetes, Amazon Web Services (AWS) and Istio, made available via the repository, which Chakravarty notes will increase the confidence IT teams have in those integrations.
Developed under the auspices of the Cloud Native Computing Foundation (CNCF), Prometheus is becoming a de facto monitoring tool standard as DevOps teams that have adopted Kubernetes to deploy containerized applications continue to drive adoption. A recent report published by Sysdig suggests 46% of organizations using the company’s monitoring services now also employ Prometheus.
Each instance of a Kubernetes cluster, however, will have its own instance of Prometheus. Over time, that becomes cumbersome when IT operations teams attempt to correlate data across multiple application environments, notes Chakravarty. The Sysdig platform can process more than 100 million metrics per second and will retain up to 13 months of Prometheus data. Most local instances of Prometheus only retain data for weeks, he notes.
The Sysdig platform also provides dashboards that can be immediately accessed versus waiting on a DevOps team to configure custom dashboards based on Grafana, adds Chakravarty.
The Prometheus issue is coming to a head inside many IT organizations as they struggle to operationalize Kubernetes clusters at scale. It’s one thing to employ an open source tool to monitor a handful of Kubernetes clusters. However, once hundreds of clusters running containerized applications take their place alongside legacy monolithic applications, the tools to monitor those environments need to be able to scale. Because of that issue, it’s not uncommon for IT teams to count on tools delivered as a cloud service designed to scale up and down as required.
There may come a day when Prometheus winds up being the dominant monitoring platform employed in the enterprise. Given how far off that day really is, however, most IT organizations for the foreseeable future will be applying a hybrid approach to monitoring.