How to Master Multi‑Cloud Application Delivery with OAM and AppManager
This talk explains the challenges of managing applications across fragmented multi‑cloud Kubernetes environments and shows how the Open Application Model (OAM) and Alibaba's AppManager provide a cloud‑native, vendor‑agnostic solution for efficient delivery, governance, and unified control.
Introduction
Guo Yaoxing from Alibaba Cloud Big Data Infrastructure shares experience on managing and delivering applications in a fragmented multi‑cloud Kubernetes environment.
Pain Points in Multi‑Cloud Application Management
Kubernetes base‑plate adaptation across heterogeneous clusters.
Conflicts between development and operations demands.
Division of responsibilities among developers, SRE, and infrastructure teams.
Why OAM?
Open Application Model (OAM) decouples application delivery from underlying platforms, providing application‑first design, clear extensibility, and cloud‑vendor independence.
Key OAM Concepts
An Application consists of one or more Components . Each Component contains a Workload (the runtime) and optional Traits (operational attributes such as storage, networking, resource limits). Workflow and Policy add process‑level controls like approvals, rollbacks, and multi‑cluster releases.
Roles
Application developers define Components.
Application SREs define Traits, bind them with Policies and Workflows, and generate the final Application YAML.
Infrastructure operators maintain the set of Workloads and ensure platform stability.
Practical Implementation – AppManager
Alibaba built a Java‑based OAM runtime called AppManager, offering capabilities such as build, deployment, artifact management, workflow engine, plugins, multi‑cloud and multi‑environment support, and status awareness.
Scenarios
Private‑cloud deployments to isolated Alibaba Cloud K8s clusters.
Public‑cloud deployments to ACK clusters across regions.
Internal corporate clusters and Alibaba internal OXS clusters.
Open‑source community (SREWorks) supporting user‑managed clusters.
Delivery Process
Developers push code to a repository; AppManager builds Docker images and packages them as component bundles. SREs select bundles, configure target clusters, add Traits, Policies, and Workflows, producing an Application YAML that AppManager reconciles to the desired state.
Unified Control Plane
A central AppManager instance coordinates multiple isolated units, each running its own AppManager to manage local applications, enabling consistent control across heterogeneous environments.
The source code of SREWorks is open on GitHub: https://github.com/alibaba/sreworks.
Signed-in readers can open the original source through BestHub's protected redirect.
This article has been distilled and summarized from source material, then republished for learning and reference. If you believe it infringes your rights, please contactand we will review it promptly.
Alibaba Cloud Big Data AI Platform
The Alibaba Cloud Big Data AI Platform builds on Alibaba’s leading cloud infrastructure, big‑data and AI engineering capabilities, scenario algorithms, and extensive industry experience to offer enterprises and developers a one‑stop, cloud‑native big‑data and AI capability suite. It boosts AI development efficiency, enables large‑scale AI deployment across industries, and drives business value.
How this landed with the community
Was this worth your time?
0 Comments
Thoughtful readers leave field notes, pushback, and hard-won operational detail here.