跳至主要内容 跳至导航 跳至页脚
多云架构

AWS 宕机时，你不会

你的基础设施运行在多个云上。当一个故障时，其他的自动接管。

每朵云都有备份。但没有一个能在几分钟内故障转移到另一朵云。Rediacc 可以。

免费开始使用社区版 查看故障转移演示 → →
$ rdc repo sync push-all
✓ Computing incremental diff .... done
✓ Transferring delta (2.4 GB of 380 GB) .. done
✓ Applying snapshot on secondary .. done
✓ Verifying 8 containers ........ 8/8 healthy
✓ Replication complete: secondary-dc in sync
⚠ Alert: primary health check failed (us-east-1 unreachable)
✓ Failover initiated ............ done
✓ DNS updated to secondary-dc ... propagated
Delta: 2.4 GB    Failover: 22s    Services: 8/8 online    Downtime: 0

输出仅作示例，实际运行可能有更多日志。 CLI reference: rdc repo sync

<30s
Failover Time
99.99%
Data Integrity
3+
Cloud Providers Supported
问题所在

单云就是单点故障

2024 年关键云故障事件同比增长 18%，持续时间延长 18.7%，其中 68% 由人为错误导致 [2]。当你唯一的云宕机时，一切都宕机了。你的客户、你的收入、你的声誉——全部沦为他人基础设施的人质。

$300K+ per hour of downtime for 90% of mid-to-large enterprises ITIC 2024 Hourly Cost of Downtime Survey [1]
18% increase in critical cloud outage events in 2024 Parametrix 2024 Cloud Outage Risk Report [2]
$400B annual downtime cost for Global 2000 companies Splunk / Oxford Economics 2024 [3]
传统方式
Min 0 Outage detected
Min 15 Assess damage
Hour 1 Manual failover
Hour 4 DNS propagation
Hour 8+ Still degraded
使用 Rediacc
Detect
<30s
Online
真实成本

一次云故障花费你多少钱？

拖动滑块以匹配你的业务。查看单云依赖的真实成本。

计算你的故障风险敞口

不使用 Rediacc
客户流失（预估）$16,000
故障转移人力成本$4,800
因故障损失的收入$80,000
SLA 罚款风险$8,000
年度故障成本
$108,800
使用 Rediacc
客户流失（预估）$0
故障转移人力成本$0
因故障损失的收入$0
SLA 罚款风险$0
年度故障成本
$0
收入损失：小时收入 × 故障小时数。故障转移人力：3 名工程师 × 故障转移小时数 × $200/小时 × 事故次数。客户流失：直接收入损失的 20%（保守估计）。SLA 罚款：收入损失的 10%。ITIC 2024 发现 41% 的企业报告每小时停机成本为 $1M-$5M [1]。Rediacc 的 <30 秒自动故障转移消除了长时间停机窗口。
工作原理

一条命令，始终在线。

1

Replicate

Run rdc replicate production --target secondary. Incremental sync — only changed blocks transfer.

2

Monitor

Continuous health checks detect failures instantly. Connects to your monitoring stack (Grafana, Prometheus).

3

Failover

Traffic shifts to healthy infrastructure in under 30 seconds. Automatic DNS updates. Zero manual steps.

主节点 (us-east-1) 宕机
nginx :443
api-server :8080
postgres :5432
redis :6379
8 个容器 不可达
故障转移
22s
备节点 (eu-west-1) 活跃
nginx :443
api-server :8080
postgres :5432
redis :6379
8 个容器 380 GB
底层技术

为什么跨云复制既快又便宜

Rediacc 使用 btrfs send/receive 在文件系统层面计算增量差异。不是传输完整镜像，只有变更的数据块在站点之间移动。一个 380 GB 的环境如果每天有 2.4 GB 的变更，就只传输 2.4 GB——而非 380 GB。这意味着 98% 的带宽节省。复制可以持续运行或按计划执行。

传统方式
Rediacc (btrfs send/receive)
Full VM image transfer: 380 GB per sync cycle, hours to complete
Incremental diff: only changed blocks, 2.4 GB typical daily transfer
Vendor-locked replication (AWS to AWS, Azure to Azure)
Cloud-agnostic: any Linux machine to any Linux machine
Manual DNS failover: 15-60 minutes of engineer time
Automated health check + DNS failover in under 30 seconds
Egress fees: $0.09/GB, ~$34/sync for 380 GB full image
Egress fees: $0.09/GB, ~$0.22/sync for 2.4 GB incremental diff
重要意义

您将获得什么

Always online

One cloud fails, another takes over. No manual intervention. Automatic failover triggers on health check failure. Connects to your monitoring stack.

Multi-cloud by default

AWS, Azure, GCP, Hetzner, or your own hardware. Mix and match. Any Linux machine with btrfs can be a replication target.

No vendor lock-in

Move between clouds whenever you want. Open-format snapshots. No proprietary agents. Standard btrfs — available in every modern Linux kernel.

差距对比

没有其他工具能在数秒内完成跨云故障转移

云原生工具在其云内复制。备份工具在数小时后恢复。Rediacc 跨任何提供商复制，并在 30 秒内故障转移。

功能 AWS BackupVeeamZertoDruva Rediacc
Cross-cloud replication ✗ ✓ ✓ ✗ ✓
Failover under 30 seconds ✗ ✗ ✗ ✗ ✓
Incremental-only transfer (98% savings) ✗ ✓ ✓ ✗ ✓
Any-to-any provider failover ✗ ✗ ✗ ✗ ✓
Self-hosted, no cloud agent ✗ ✓ ✗ ✗ ✓
Automatic DNS failover ✗ ✗ ✓ ✗ ✓
Full-stack replication (app + DB) ✗ ✓ ✓ ✗ ✓
在去年三月的 AWS us-east-1 故障期间，我们的 12 个生产服务在 22 秒内故障转移到了备用数据中心。我们的客户毫无察觉。我们的竞争对手宕机了 6 小时。那周我们从他们的客户那里获得了三个新的企业合同。
→
故障转移速度快 982 倍

不要把业务押注在一朵云上

从免费的社区版开始。免费跨地域复制。无需信用卡。

免费开始使用社区版 最多 10 个工作负载永久免费
$ rdc repo push --to secondary-dc -m primary
复制任何容器化工作负载
数据库、CI/CD、CMS、监控、认证、存储——只要在容器中运行，Rediacc 就能跨云复制它。

探索其他解决方案

multi cloud

Cloud Outage Protection

When AWS goes down, you don’t

Current page multi cloud

Vendor Lock-in Freedom

Your infrastructure. Your choice. Always.

→ multi cloud

Failover Testing

Test your failover before you need it

→ ransomware

Immutable Backups

Backups that ransomware can’t touch

→ backups

Backup Verification

Every backup verified automatically

→ encryption

Encryption You Control

Your keys. Your encryption. No exceptions.

→ dev env

Environment Cloning

Clone production in 60 seconds

→ defense

AI Pentesting

Clone production. Let AI attack it.

→
来源与参考
  1. ITIC, “2024 Hourly Cost of Downtime Report,” 2024. “The average cost of a single hour of downtime now exceeds $300,000 for over 90% of mid-size and large enterprises.” itic-corp.com
  2. Parametrix, “2024 Cloud Outage Risk Report,” 2024. “Critical cloud service interruption events increased by 18.0% and lasted 18.7% longer in 2024 than in 2023. Human error accounted for 68% of incidents.” www.reinsurancene.ws
  3. Splunk & Oxford Economics, “The Hidden Costs of Downtime,” 2024. “The total cost of downtime for Global 2000 companies is $400B annually, or 9 percent of profits, when digital environments fail unexpectedly.” www.splunk.com
产品性能声明基于 Rediacc 的 btrfs 写时复制架构。计算器估算使用行业标准成本模型；实际成本因组织而异。