Stan the SRE : “Causal AI speeds incident remediation”

Accident – Death – Obituary News : : 1. Incident Remediation Efficiency
2. Causal AI for Root Cause Analysis

Learn how the new feature “Probable Root Cause” from Instana® is revolutionizing incident remediation for businesses. With application outages costing up to USD 500,000 per hour, Site Reliability Engineers (SREs) often spend hours identifying and resolving issues. Instana’s Causal AI analyzes call statistics, topology, and surrounding data to quickly pinpoint the probable source of failure, saving time and costs for businesses. By combining disparate data sources with causal AI, Instana provides detailed and accurate root cause analysis, even in cases with limited data granularity. Explore the future potential of enhanced explainability and actionable recommendations for intelligent incident remediation. Discover more about IBM Instana’s capabilities and the intelligent remediation pipeline.

In the fast-paced world of business applications, downtime is a costly affair. With the increasing complexity of applications and the shift towards digitization, the estimated cost of an average outage can range from USD 50,000 to 500,000 per hour. This is where the new feature, Probable Root Cause, comes into play as part of Intelligent Incident Remediation from Instana®.

**The Data**

Instana sets itself apart by monitoring 100% of every call trace, providing detailed information about infrastructure and application activities. With one-second granularity metrics, events, and a dynamic application and infrastructure topology, Instana offers unparalleled data availability. This allows for precise identification of probable root causes using causal AI.

**The Assumptions**

Traditional IT management tools often struggle with complete and granular application topology. Due to specialized processes and team ownership, many tools lack full observability into multi-layered applications. However, Instana’s use of causal AI and a versatile algorithm enables the identification of root causes even with limited data granularity and partial topology.

**The Method**

By leveraging causal AI, Instana combines data from various sources such as calls, metrics, events, and topology to pinpoint root causes of application faults. This approach not only identifies problematic components but also provides insights into why these entities were flagged as probable causes, instilling confidence and trust in the remediation process.

**An Example Use Case with Stan the SRE**

Let’s follow Stan, an SRE at a small company monitoring the robot-shop application with Instana. Upon receiving an alert about a performance issue, Stan utilizes the Probable Root Cause feature to quickly identify the root cause—a specific process within the catalogue service. By analyzing call traces and infrastructure metrics, Stan efficiently resolves the issue, saving valuable time and resources in the incident investigation and remediation workflow.

**A Vision for the Future**

In the coming months, Instana plans to enhance its root causing capabilities further. With a focus on enhanced explainability and a deeper analysis of what went wrong, the goal is to provide SREs with detailed insights and action recommendations for remediation. This ongoing collaboration between engineering and IBM® research showcases the potential for intelligent incident remediation to evolve and adapt to meet the needs of modern businesses.

The Probable Root Cause Feature, currently in tech preview, demonstrates the commitment to innovation and problem-solving. As businesses continue to navigate the complexities of digital transformation, tools like Instana’s Intelligent Incident Remediation offer a glimpse into the future of efficient and effective incident resolution. Stay tuned for the full version of this feature, coming soon!

**Learn more about IBM Instana’s probable root cause capabilities and the intelligent remediation pipeline**