Announcing exciting new capabilities for IBM Cloud Monitoring with Sysdig.
Today, we are announcing a major step forward to IBM Cloud Monitoring with Sysdig—cloud-scale Prometheus compatibility, first-class support for the Prometheus query language (PromQL), and an enhanced user experience that makes it even easier to get up and running with monitoring the health and performance of your apps, infrastructure, and IBM services.
Using Sysdig Monitor for Prometheus provides a complete end-to-end solution with open source compatible metric collection, flexibility to query metrics using either PromQL or our simple form-based querying. Additionally, ensure regulatory compliance with enterprise-grade access controls via enhanced Teams support and RBAC.
Over the course of the next several weeks, customers will receive access to these features and more.
So what can customers expect to see with this new rollout?
Native Prometheus metric ingestion
We have embedded a lightweight Prometheus server directly into the Sysdig agent to facilitate metric collection and forwarding to the Sysdig metrics store. This also supports filtering and relabeling using Prometheus syntax. Now, you can use the hundreds of available Prometheus exporters with Sysdig in the same way that you would use them within the open source ecosystem, including the Sysdig catalog of supported Prometheus monitoring integrations through PromCat.
PromQL compatibility
PromQL is a powerful way to query your metrics within Sysdig; you can perform complex mathematical operations, statistical analyses, and use a variety of functions to dig deeper with metrics. Using PromQL, you’ll now be able to answer more questions about the health and performance of your infrastructure using advanced functions and operators in both Dashboards and Alerts.
While mastering PromQL can make it feel like you’ve leveled up your monitoring expertise, it does have a steep learning curve, which is something we didn’t want to overlook. We’ve ensured that the form-based dashboard editor is retained for users wanting to get up and running quickly. If you want to run a basic query to have a look at your CPU usage grouped by each Kubernetes deployment, you shouldn’t have to write complex PromQL queries composed of joins and functions. And it shouldn’t be complex for non-technical folks who just want to run a simple report to perform rightsizing tasks.
Alongside Prometheus compatibility, the new metrics store offers increased long-term retention—10s interval data will now be kept for four days (previously, four hours) and one-minute interval data will now be kept for two weeks. Further roll-ups will, ultimately, provide 13 months of data that can be used for trend analysis and reporting.
New dashboards
We’ve introduced a more fluid, natural dashboard-building experience with a complete redesign of the UI, including a new panel editor that makes crafting the best way to visualize your metrics a lot easier. Other usability enhancements include multi-metric and multi-segmentation panels, improvements to the visualization of different panels, Teams-sharing capabilities and a PromQL mode to query Prometheus metrics.
The good news is that all of your dashboards will be migrated into the new format for you—there’s nothing you need to do.
Read more about Sysdig new dashboards: New and improved dashboards: PromQL, Teams sharing, and more!
Improved sharing capabilities
Sysdig Teams allow portions of your organization to only access the Prometheus metrics and telemetry that they care about. With full RBAC support, you can provide an application team responsible for maintaining an analytics tooling system access to only the metrics being emitted from their namespace. You can also give an on-call team read-only access to production hosts.
Starting today, you can share your dashboard with users within your Sysdig Team or share it across Teams with fine-grained access controls. Define who should be able to see those dashboards and what level of access they should be granted (View Only or Collaborator with edit privileges).
Beginning this week, we will deploy this code to the Sydney and Tokyo multi-zone regions, with other regions to soon follow in order to accomplish general availability in all six multi-zone regions by mid-August 2020.
We’d love your feedback so feel free to email us directly to continue the conversation and improvement of this offering!