Virtualization Technology News and Information
Chronosphere Connects Traces with Metrics in its Monitoring Platform to Reduce Business Downtime

Chronosphere, the company pushing observability and cloud native monitoring to new levels, announced that it has added tracing capabilities to its platform for monitoring today's complex technology stacks at large scale. This allows users to correlate traces with metrics, adding contextual information to alerts and dashboards in order to reduce the amount of time it takes to resolve production issues. The new feature is based on open source and CNCF technologies including OpenTelemetry, OpenMetrics and M3, all three of which Chronosphere contributes to.

Chronosphere, founded by former Uber engineers who created M3, formally launched earlier this month. Dennis Drogseth, vice president at Enterprise Management Associates (EMA), said: "Chronosphere, born out of demanding real-world requirements, is a breath of fresh air with its fast time to value and proven capabilities in helping large, enterprise IT organizations proactively address complex service delivery problems across multi-cloud and hybrid cloud environments."

Many of today's monitoring platforms offer both metrics and tracing capabilities and claim tight integrations between the two observability pillars. However diving into these integrations, it becomes apparent that they are two separate systems that are loosely linked via a common period of time and a common set of attributes and labels. While this does help narrow down the search space, manual intervention is often required to link and make sense of the two types of monitoring data.

The problem with manual intervention is that each minute the issue remains unresolved, the business is negatively impacted and due to this, engineers are often under immense pressure to find the cause of issues quickly. An ideal solution would present all the context of the issue upfront, reducing the time it takes to identify and mitigate issues - and that is what Chronosphere's monitoring platform offers. 

Chronosphere's solution to this problem is to deeply link the traces and metrics from the moment they are emitted. The trace identifiers are stored right next to the metric values in Chronosphere's metrics storage system M3. When a user receives an alert or views their dashboard, the very traces that constituted those metric values are displayed instantly. This saves valuable minutes during an outage and helps users identify and mitigate issues much faster.

"The best part of our solution is that we can achieve this deep linking in a cost-efficient way," said Rob Skillington, co-founder and CTO of Chronosphere. "This would be an easy problem to solve if you simply stored every trace, but in most cases, that's way too expensive to do at any real level of scale."

Chronosphere achieves cost efficiency by leveraging their existing metrics aggregation tier to selectively store unique traces. This allows users to both find that needle in a haystack - the one in a thousand error or slow request - while using a fraction of the storage. 

Chronosphere will be showcasing its new tracing feature and platform, currently in private beta, this week at its booth at KubeCon + CloudNativeCon North America. Chronosphere's Skillington will give a talk titled ‘Deep Linking Metrics and Traces with OpenTelemetry, OpenMetrics and M3' at the conference on Thursday at 10:55 a.m.

Published Wednesday, November 20, 2019 6:35 AM by David Marshall
Filed under:
There are no comments for this post.
To post a comment, you must be a registered user. Registration is free and easy! Sign up now!
<November 2019>