aqstack/sentinel

Self-healing edge computing agent with predictive failure detection and partition-resilient orchestration for Kubernetes

54
/ 100
Established

This project helps operations engineers and site reliability engineers manage fleets of Kubernetes edge devices. It monitors critical system health indicators like CPU temperature and memory pressure to predict hardware failures before they occur. It then automatically takes action to move workloads, preventing outages even when disconnected from the central control plane, providing robust, self-healing edge operations.

373 stars.

Use this if you manage Kubernetes clusters deployed in challenging edge environments with unreliable network connectivity or limited resources, and need proactive failure prevention and autonomous operation.

Not ideal if your Kubernetes clusters are exclusively in stable cloud or datacenter environments with constant control plane access and abundant resources.

edge-computing kubernetes-operations predictive-maintenance distributed-systems site-reliability
No Package No Dependents
Maintenance 6 / 25
Adoption 10 / 25
Maturity 13 / 25
Community 25 / 25

How are scores calculated?

Stars

373

Forks

340

Language

Go

License

MIT

Last pushed

Jan 04, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/mlops/aqstack/sentinel"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.