AWS DevOps Guru is a fully managed AIOps service that:<\/p>\n
- \n
- Continuously analyzes CloudWatch metrics<\/li>\n
- Correlates operational signals<\/li>\n
- Detects anomalous behavior<\/li>\n
- Identifies related impacted resources<\/li>\n
- Provides actionable recommendations<\/li>\n<\/ul>\n
It builds machine learning baselines using historical operational data and identifies deviations that may indicate risk, degradation, or failure.<\/p>\n
Unlike basic monitoring, it focuses on system-level insights, not just metric alerts.<\/p>\n
DevOps Guru analyzes telemetry from CloudWatch \u2014 metrics, logs, configuration changes \u2014 and builds a behavioral baseline over time. It learns what \u201cnormal\u201d looks like for your environment.<\/p>\n
When deviations occur, it doesn\u2019t just flag a single metric. It tries to group related anomalies together and present them as a single insight.<\/p>\n
![\"AWS](\"https:\/\/www.tothenew.com\/blog\/wp-ttn-blog\/uploads\/2026\/03\/image-1.png\")
AWS DevOps Guru<\/p><\/div>\n
DevOps Guru follows the following workflow:<\/strong><\/p>\n
1. Monitor CloudWatch Metrics and Logs<\/strong>
\nContinuous ingestion of operational telemetry.<\/p>\n2. Apply Machine Learning Models<\/strong>
\nBuild baselines and detect anomalies in behavior.<\/p>\n3. Correlate with Events and Resource Relationships<\/strong>
\nAnalyze configuration changes, deployments, and service dependencies.<\/p>\n4. Construct Recommendations<\/strong>
\nUse learned patterns and AWS best practices to suggest corrective actions.<\/p>\n5. Generate Insights<\/strong>
\nPresent a consolidated, contextual operational view.<\/p>\nThis workflow shifts monitoring from simple detection to contextual analysis.<\/p>\n
![\"Workflow\"](\"https:\/\/www.tothenew.com\/blog\/wp-ttn-blog\/uploads\/2026\/03\/image-9-1024x683.png\")
Workflow<\/p><\/div>\n
Set up of\u00a0 AWS DevOps Guru :\u00a0<\/strong><\/h2>\n
Step 1: Open the AWS Console and search for DevOps Guru<\/h3>\n
![\"image\"](\"https:\/\/www.tothenew.com\/blog\/wp-ttn-blog\/uploads\/2026\/03\/image-3.png\")
image<\/p><\/div>\n
Step 2 : Enabling DevOps Guru<\/h3>\n
Choose analysis coverage, SNS topic and enable the service.<\/p>\n
![\"image\"](\"https:\/\/www.tothenew.com\/blog\/wp-ttn-blog\/uploads\/2026\/03\/image-5-1.png\")
image<\/p><\/div>\n
<\/p>\n
![\"Dashboard\"](\"https:\/\/www.tothenew.com\/blog\/wp-ttn-blog\/uploads\/2026\/03\/image-5-2.png\")
Dashboard<\/p><\/div>\n
<\/h2>\n
<\/h2>\n
AWS DevOps Guru Insights<\/strong><\/h2>\n
Amazon DevOps Guru generates an insight upon detection of anomalous behaviour in the operational applications. It contains a list of the metrics and a list of the events that were used to identify the unusual behaviour. Along with this, the insight contains one or more recommendations to mitigate the issue.<\/p>\n
There are two insight types –<\/p>\n
Reactive insights<\/strong> – Reactive insights have recommendations you can take to address issues that are happening now.<\/p>\n
Proactive insights –<\/strong> Proactive insights have recommendations that address issues that DevOps Guru predicts will occur in the future.<\/p>\n
![\"Insights\"](\"https:\/\/www.tothenew.com\/blog\/wp-ttn-blog\/uploads\/2026\/03\/image-6.png\")
Insights<\/p><\/div>\n
<\/h3>\n
<\/h2>\n
Choosing Resources to Analyze<\/h2>\n
When enabling DevOps Guru, AWS asks you to define the analysis scope. This determines which resources DevOps Guru will monitor, model, and generate insights.<\/p>\n
![\"Resources\"](\"https:\/\/www.tothenew.com\/blog\/wp-ttn-blog\/uploads\/2026\/03\/image-7.png\")
Resources<\/p><\/div>\n
![\"Resources\"](\"https:\/\/www.tothenew.com\/blog\/wp-ttn-blog\/uploads\/2026\/03\/image-8.png\")
Resources<\/p><\/div>\n
<\/h2>\n
<\/h2>\n
CloudWatch Anomaly Detection vs AWS DevOps Guru<\/h2>\n
CloudWatch Anomaly Detection and AWS DevOps Guru both use machine learning, but they operate at different levels. CloudWatch works on individual metrics, building a baseline and alerting when that specific metric behaves unusually. It improves traditional threshold-based monitoring by making it adaptive.<\/p>\n
DevOps Guru takes a broader approach. It analyzes multiple metrics, logs, and resource relationships together, grouping related anomalies into a single insight. Instead of just detecting abnormal behavior, it helps identify where the issue might be coming from.<\/p>\n
AWS DevOps Guru Use cases<\/h2>\n
- \n
- Scale and maintain availability in complex applications<\/li>\n
- Proactively identify resource limits (e.g., memory, CPU, disk)<\/li>\n
- Detect abnormal application behavior using machine learning models<\/li>\n
- Use ML models to limit alarm noise and focus on critical issues<\/li>\n
- Consolidate operational data from multiple sources for unified insights<\/li>\n
- Support insight creation for both reactive and proactive operational events<\/li>\n
- Provide early warning and proactive insights before issues impact customers<\/li>\n<\/ul>\n
Conclusion<\/h2>\n
AWS DevOps Guru adds intelligence on top of traditional monitoring by correlating anomalies across services and presenting them as meaningful insights. It is most valuable in complex production environments where multiple components interact and manual investigation becomes time-consuming.<\/p>\n","protected":false},"excerpt":{"rendered":"
Introduction Cloud monitoring has evolved over the years and we have moved from manual static monitoring of thresholds to dynamic anomaly monitoring, AI and ML-based operational tasks.Here AWS DevOps Guru comes into the picture as an Mananged machine learning service in Cloud Operations. AWS DevOps Guru is an AIOps solution that can detect operational anomalies […]<\/p>\n","protected":false},"author":1863,"featured_media":0,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"iawp_total_views":22},"categories":[5877],"tags":[248,1916,7501],"aioseo_notices":[],"_links":{"self":[{"href":"https:\/\/www.tothenew.com\/blog\/wp-json\/wp\/v2\/posts\/78097"}],"collection":[{"href":"https:\/\/www.tothenew.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.tothenew.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.tothenew.com\/blog\/wp-json\/wp\/v2\/users\/1863"}],"replies":[{"embeddable":true,"href":"https:\/\/www.tothenew.com\/blog\/wp-json\/wp\/v2\/comments?post=78097"}],"version-history":[{"count":13,"href":"https:\/\/www.tothenew.com\/blog\/wp-json\/wp\/v2\/posts\/78097\/revisions"}],"predecessor-version":[{"id":78119,"href":"https:\/\/www.tothenew.com\/blog\/wp-json\/wp\/v2\/posts\/78097\/revisions\/78119"}],"wp:attachment":[{"href":"https:\/\/www.tothenew.com\/blog\/wp-json\/wp\/v2\/media?parent=78097"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.tothenew.com\/blog\/wp-json\/wp\/v2\/categories?post=78097"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.tothenew.com\/blog\/wp-json\/wp\/v2\/tags?post=78097"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}