Managing Distributed ECS Resources with ACK Edge and Kubernetes

ACK Edge provides a standard, secure, and highly available Kubernetes cluster for distributed computing scenarios, allowing geographically dispersed ECS resources to be integrated into a single cloud‑native cluster for unified lifecycle management and resource scheduling.

Scenario Description

Users often have ECS instances spread across multiple VPCs, regions, or accounts and need a single Kubernetes cluster to manage these resources and the applications running on them.

Solution Advantages

Standard cloud‑native interfaces reduce operational costs.

Alibaba Cloud‑managed control plane with SLA, eliminating the need to operate the Kubernetes control plane.

Seamless integration with existing cloud services (elastic compute, networking, storage, observability) to ensure stable application operation.

Support for dozens of heterogeneous operating systems.

Edge autonomy, cloud‑edge operation channels, and modular management for centralized‑edge scenarios.

Optimized cloud‑edge traffic reduces bandwidth costs; a single cluster can manage thousands of nodes.

Solution Examples

Example 1: Managing Region‑Distributed Applications

When many ECS instances are scattered across regions, create an ACK Edge cluster and use DaemonSet or OpenKruise DaemonSet to deploy and manage containers uniformly. Typical use cases include security agents, distributed load testing, and cache acceleration services.

Example 2: Single‑Region GPU Shortage

If a region lacks GPU resources for AI tasks, purchase GPU instances in another region, add them to the ACK Edge cluster, and let the scheduler place workloads on the newly added GPU nodes.

Operational Steps

1. Environment Preparation

Select a central region and create an ACK Edge cluster.

Install OpenKruise via the component management console.

Create edge node pools for each region and attach the corresponding ECS instances.

2. Deploy Business Workloads Using Native DaemonSet

In the cluster console, navigate to the DaemonSet page, choose the namespace and deployment method, and follow the prompts to complete deployment.

For upgrades, edit the DaemonSet template to update version and configuration.

3. Deploy Business Workloads Using OpenKruise DaemonSet

In the cluster console, go to the workload page, select the YAML deployment option, paste the customized YAML, and submit.

For upgrades, edit the YAML of the OpenKruise DaemonSet directly.

Sample YAML for a TensorFlow Deployment

apiVersion: apps/v1
kind: Deployment
metadata:
  name: tensorflow-mnist
  labels:
    app: tensorflow-mnist
spec:
  replicas: 1
  selector:
    matchLabels:
      app: tensorflow-mnist
  template:
    metadata:
      name: tensorflow-mnist
      labels:
        app: tensorflow-mnist
    spec:
      containers:
      - name: tensorflow-mnist
        image: registry.cn-beijing.aliyuncs.com/acs/tensorflow-mnist-sample:v1.5
        command:
        - python
        - tensorflow-sample-code/tfjob/docker/mnist/main.py
        - --max_steps=100000
        - --data_dir=tensorflow-sample-code/data
        resources:
          limits:
            nvidia.com/gpu: "1"
          requests:
            nvidia.com/gpu: "1"
        workingDir: /root

Related Documentation

Creating an ACK Edge Cluster

Managing Edge Node Pools

ACK Edge Component Overview

ACK Edge Billing

OpenKruise Advanced DaemonSet