Managing Containerized Software on Edge Computers with Open Horizon
summary
Open Horizon is an IBM-backed fleet manager that installs, updates and secures container images and data files on heterogeneous edge nodes, supporting Kubernetes, Docker/Podman and architectures including x86, ARM, Power and RISC‑V. It runs a small autonomous agent (~30 MB agent runtime) with a zero trust model, separate model-management for live ML models, HashiCorp Vault for secrets, SDO/FDO onboarding, and is tested at scale (40,000 devices on one hub).
AI-written summary · about
chapters (16)
- 0:00Intro
- 1:10What is Open Horizon?
- 8:09The Genesis of Open Horizon
- 9:18Challenges on the edge...
- 10:06Open Horizon Tenets
- 12:28Open Horizon Architecture Overview
- 18:07Edge Computing Examples....
- 18:491: Open Horizon Agent
- 21:442: Open Horizon Management Hub
- 25:49Architecture - How does it work?
- 36:55Architecture - Summary
- 37:28Open Horizon Policies
- 39:51Open Horizon Usage
- 41:51Open Horizon and Software Bill of Materials (SBOM)
- 42:50Open Horizon: Videos, Docs, Example Code...
- 43:08Open Horizon: Contact Us...
related talks
36mOpenEdge, the General Framework of Edge Computing
10mDevelop once, deploy anywhere: flexible infrastructure extends to the edge
39mIoT Application Running on KubeEdge + ARM Platform
10mKubernetes: Enabling Edge Workloads
24mDemo of Kubernetes at the Edge for AI Model Deployments
39mKubernetes IoT Edge Working Group: Edge Device Onboa...
12mSMARTER: Experiences with Cloud Native on the Edge
39mIoT Application Running on KubeEdge + ARM Platform