MIDAOPS02-BP01 Use enhanced monitoring and alerting in the cloud and at the edge
Comprehensive observability is crucial for manufacturing teams to proactively monitor the health and performance of the data infrastructure.
Desired outcome: Achieve real-time visibility into the entire industrial data infrastructure, enabling proactive issue detection, rapid troubleshooting, and maintenance of optimal system performance across both cloud and edge environments.
Benefits of establishing this best practice: By implementing this robust observability framework, a manufacturing organization can gain visibility and insight needed to proactively manage the reliability, performance, and security of their data-driven operations. The observability tools and practices enable the team to quickly detect, investigate, and resolve issues before they impact production.
-
Enables proactive identification of potential issues before they impact operations
-
Reduces mean time to detection (MTTD) and resolution (MTTR)
-
Provides comprehensive visibility across the entire data infrastructure
-
Facilitates data-driven decision making for system optimization
-
Enhances overall system reliability and performance
Level of risk exposed if this best practice is not established: High
Implementation guidance
Establish a comprehensive monitoring and alerting framework that covers both cloud and edge components of your industrial data infrastructure.
Implementation steps
-
Use AWS AWS IoT Greengrass to collect and stream operational telemetry data, such as device health metrics and diagnostic logs, back to the cloud. This allows the manufacturing team to maintain visibility into the state of edge components, even in disconnected scenarios.
-
Use Amazon CloudWatch to collect and analyze metrics, logs, and events from across the hybrid edge-cloud architecture. This includes metrics like data ingestion rates, database performance, and resource utilization. CloudWatch dashboards and alarms provide visibility into the overall system health, allowing the team to quickly identify and address issues. Set baseline performance metrics and configure automated alerting thresholds.
-
To enable end-to-end tracing and diagnostics, use AWS X-Ray. X-Ray provides detailed insights into the performance and dependencies of distributed components, helping the team rapidly pinpoint the root cause of problems.
-
Consider recording all API calls and configuration changes using AWS CloudTrail, enabling comprehensive auditing and security monitoring of the data infrastructure.
Key AWS services
-
AWS CloudTrail
-
Amazon CloudWatch
-
AWS AWS IoT Greengrass
-
AWS X-Ray
Resources
Related documents:
