Edge-to-cloud monitoring solution for industrial IoT environments
Rekola, Miska (2026)
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-2026052717610
https://urn.fi/URN:NBN:fi:amk-2026052717610
Tiivistelmä
The aim of this thesis was to develop a centralized monitoring and observability solution for the industrial IoT gateways within the Tamtron mScales ecosystem. The mScales platform operates a distributed fleet of Linux-based gateway devices deployed across vehicle scale sites, and the absence of a dedicated system-health monitoring channel meant that hardware issues could go undetected until they directly affected the weighing service.
To address this gap, a telemetry pipeline was designed to collect device-level performance metrics at the edge and transmit them to the cloud. AWS IoT Greengrass forwards structured log data formatted in the Embedded Metric Format to AWS CloudWatch, which automatically extracts the embedded metrics while retaining the raw logs for troubleshooting. The data is then forwarded to Datadog via a Lambda-based function, where custom dashboards and alert monitors provide fleet-wide visibility, with notifications delivered to the support team through a Microsoft Teams webhook.
The resulting solution gives administrators continuous visibility into gateway performance and enables issues to be identified before they cause production disruptions, while also establishing the technical groundwork for future condition-based predictive maintenance.
To address this gap, a telemetry pipeline was designed to collect device-level performance metrics at the edge and transmit them to the cloud. AWS IoT Greengrass forwards structured log data formatted in the Embedded Metric Format to AWS CloudWatch, which automatically extracts the embedded metrics while retaining the raw logs for troubleshooting. The data is then forwarded to Datadog via a Lambda-based function, where custom dashboards and alert monitors provide fleet-wide visibility, with notifications delivered to the support team through a Microsoft Teams webhook.
The resulting solution gives administrators continuous visibility into gateway performance and enables issues to be identified before they cause production disruptions, while also establishing the technical groundwork for future condition-based predictive maintenance.