Bowen Chen
InsightFinder is a SaaS platform that uses AI-backed predictive analytics to predict and prevent production incidents. Using InsightFinder with Datadog, you can quickly identify hidden correlations in your application metrics, logs, and events and address application issues before they devolve into production outages and create customer impact.
After you sign up for an InsightFinder trial in the Datadog Marketplace and install the out-of-the-box (OOTB) InsightFinder integration, you can begin sending metrics from Datadog to InsightFinder’s unsupervised ML engine to predict incidents and identify hidden correlations. In this post, we’ll cover how to use the InsightFinder integration to visualize a high-level overview of these predictions with an OOTB dashboard and send InsightFinder’s incident predictions to Datadog Incident Management.
After installing the InsightFinder integration, Datadog will automatically populate an OOTB dashboard with key incident data to help you visualize the severity, volume, and type of InsightFinder’s predicted incidents. Using the dashboard’s Predicted Incidents Timeline, you can track your InsightFinder incidents over time to determine whether any patterns exist. For example, if InsightFinder routinely predicts incidents at a certain time of day, it could indicate that your application struggles to handle traffic during peak hours. Since InsightFinder learns and localizes root causes in real time by using your logs, metrics, and traces, you can pinpoint the source of incidents before they generate customer impact.
Analyzing the count of predicted incidents can serve as a general indicator of the health of your application. Whether or not a predicted incident actually evolves into a production incident or outage, your metrics are still likely displaying signs of either erratic behavior or unhealthy performance. By preemptively addressing high volumes of predicted incidents, you can ensure that troublesome application code doesn’t devolve into high volumes of actual production incidents.
After you configure InsightFinder to receive metrics, logs, and events from Datadog, you can use your Datadog telemetry with InsightFinder’s unsupervised learning algorithms to begin predicting incidents and detecting anomalies. This practice enables your team to get ahead of potential outages and make adjustments before an incident occurs.
You can then go to the “Settings” tab and configure InsightFinder to send these incident predictions and detections directly to Datadog Incident Management. Viewing your InsightFinder predictions as Datadog incidents enables your team to situate these predictions within the rest of your application’s incident stream and coordinate a response by using familiar workflows. Incident responders can then investigate the predicted incident or detected anomalies within Datadog APM, where they can home in on errorful traces and identify resource bottlenecks by using the Continuous Profiler.
Using metrics, logs, and events from Datadog, InsightFinder’s integration enables you to visualize predicted incidents with an OOTB dashboard and get ahead of them before they impact customers. To begin monitoring your InsightFinder predictions with Datadog, sign up for a 14-day trial of the software license in the Datadog Marketplace and install the free integration . If you aren’t already a Datadog customer, you can learn more about the Marketplace in our blog post and sign up for a free 14-day trial of Datadog today.
The ability to promote branded marketing tools is a membership benefit offered through the Datadog Partner Network. If you’re interested in developing an integration or application that you’d like to promote, you can contact us at marketplace@datadog.com.
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。