lab_code: demo_001
title_en: "TiDB Cloud Resilience: Surviving Data Center Failures"
title_zh: "TiDB Cloud 容灾能力：应对数据中心级故障"
title_ja: "TiDB Cloud の耐障害性：データセンター障害への対応"
title_ko: "TiDB Cloud 복원력: 데이터 센터 장애 대응"
description_en: This hands-on lab explains why TiDB Cloud keeps serving applications when an entire data center or availability zone fails. The exercise runs on TiDB Self-Managed so you can inspect PD, TiKV, and TiDB nodes component by component—placement across zones, Raft replication, and client failover—while TiDB Cloud delivers the same distributed engine as a fully managed service. By simulating the loss of one availability zone and verifying that workloads continue, you connect TiDB Cloud resilience to the underlying architecture you can see in this lab.
description_zh: 本实验说明当整个数据中心或可用区发生故障时，TiDB Cloud 为何仍能继续为应用提供服务。实验在 TiDB 自管部署（TiDB Self-Managed）环境中进行，以便你逐个观察 PD、TiKV、TiDB 等组件——跨可用区部署、Raft 复制以及客户端故障转移——而 TiDB Cloud 以全托管方式提供相同的分布式引擎。通过模拟单个可用区失联并验证业务负载持续运行，你将把 TiDB Cloud 的容灾能力与实验中可见的底层架构联系起来。
description_ja: この演習では、データセンターまたはアベイラビリティゾーン全体が障害になった場合に TiDB Cloud がなぜアプリケーションを継続提供できるのかを説明します。TiDB Cloud はフルマネージドサービスであるため、本演習では PD・TiKV・TiDB をコンポーネントごとに確認できる TiDB Self-Managed 上で実施します。ゾーン間配置、Raft レプリケーション、クライアントのフェイルオーバーを観察し、1 つのアベイラビリティゾーン喪失をシミュレートしてワークロードが継続することを確認することで、TiDB Cloud の耐障害性とその基盤アーキテクチャを結び付けます。
description_ko: 이 실습은 데이터 센터 또는 가용성 영역 전체가 장애가 났을 때 TiDB Cloud가 왜 애플리케이션 서비스를 계속 제공할 수 있는지 설명합니다. TiDB Cloud는 완전 관리형 서비스이므로, 본 실습은 PD·TiKV·TiDB 노드를 구성 요소별로 살펴볼 수 있는 TiDB Self-Managed 환경에서 진행됩니다. 영역 간 배치, Raft 복제, 클라이언트 장애 조치를 관찰하고, 하나의 가용성 영역 손실을 시뮬레이션하여 워크로드가 지속되는지 확인함으로써 TiDB Cloud의 복원력과 그 기반 아키텍처를 연결합니다.
splash_image_en: https://lab-static.pingcap.com/quick-demo/demo_001_en.png
splash_image_zh: https://lab-static.pingcap.com/quick-demo/demo_001_en.png
splash_image_ja: https://lab-static.pingcap.com/quick-demo/demo_001_en.png
splash_image_ko: https://lab-static.pingcap.com/quick-demo/demo_001_en.png
lab_level: Beginner
category_en:
  - Foundational Capability
  - Disaster Recovery
category_zh:
  - 基础能力
  - 容灾
category_ja:
  - 基礎能力
  - ディザスタリカバリ
category_ko:
  - 기초 능력
  - 재해 복구
authors:
  - Todd Bao (CEG)
  - Yasuo Honda (JBG)
  - Tadatoshi Sekiguchi (JBG)
delivery_type: Self-Paced
tidb_environments:
  - TiDB Self-Managed
tidb_versions:
  - 6.5.9
  - 7.5.4
  - 8.1.1
  - 8.5.1
target_audience:
  - Architect
  - DBA
  - DevOps
regions:
  aws:
    - oregon
    - tokyo
    - singapore
    - frankfurt
credits: 1
credit_redeemable: true
employee_only: false
estimated_duration_minutes: 60

topology:
  - role: monitor
    node_count: 1
    web_ssh: true
    web_ssh_name: "TiUP Control Machine"
    ssh_user: ec2-user
  - role: pd
    node_count: 3
  - role: kv
    node_count: 3
  - role: db
    node_count: 2

requirements:
  - type: "CloudStack"

bootstrap:
  - name: "prepare monitor"
    type: "ShellScript"
    target:
      - monitor1
    script: ./scripts/prepare-monitor.sh

features:
  enable_content_explain: false