deployment

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Deployment Router

部署路由指南

Quick guide to choose the right deployment platform and skill for your workload.

这份快速指南将帮助你为工作负载选择合适的部署平台及对应技能。

Platform Decision Matrix

平台决策矩阵

Workload	Platform	Skill
Next.js / React frontend	Vercel	`vercel-deployment`
Static sites, SPAs	Vercel or Cloudflare Pages	`vercel-deployment`
FastAPI / Python backend	Fly.io	`fly-deployment`
Stateful backend (SQLite, volumes)	Fly.io	`fly-deployment`
Docker containers	Fly.io	`fly-deployment`
Object storage / CDN	Cloudflare R2	`cloudflare-r2-d1`
Edge database (SQLite)	Cloudflare D1	`cloudflare-r2-d1`
Session/config caching	Cloudflare KV	`cloudflare-r2-d1`
Serverless functions	Cloudflare Workers	`cloudflare-r2-d1`

工作负载	平台	技能
Next.js / React 前端	Vercel	`vercel-deployment`
静态站点、单页应用	Vercel 或 Cloudflare Pages	`vercel-deployment`
FastAPI / Python 后端	Fly.io	`fly-deployment`
有状态后端（SQLite、存储卷）	Fly.io	`fly-deployment`
Docker 容器	Fly.io	`fly-deployment`
对象存储 / CDN	Cloudflare R2	`cloudflare-r2-d1`
边缘数据库（SQLite）	Cloudflare D1	`cloudflare-r2-d1`
会话/配置缓存	Cloudflare KV	`cloudflare-r2-d1`
无服务器函数	Cloudflare Workers	`cloudflare-r2-d1`

Quick Reference by Stack

按技术栈快速参考

Full-Stack Next.js + Python API

全栈 Next.js + Python API

Frontend (Next.js)     → Vercel       → vercel-deployment
Backend (FastAPI)      → Fly.io       → fly-deployment
File uploads           → Cloudflare R2 or Fly volume

Frontend (Next.js)     → Vercel       → vercel-deployment
Backend (FastAPI)      → Fly.io       → fly-deployment
File uploads           → Cloudflare R2 or Fly volume

Edge-First Stack (Cloudflare)

边缘优先技术栈（Cloudflare）

Frontend (static/SPA)  → Cloudflare Pages
API (Workers)          → Cloudflare Workers  → cloudflare-r2-d1
Database               → Cloudflare D1       → cloudflare-r2-d1
Files                  → Cloudflare R2       → cloudflare-r2-d1

Frontend (static/SPA)  → Cloudflare Pages
API (Workers)          → Cloudflare Workers  → cloudflare-r2-d1
Database               → Cloudflare D1       → cloudflare-r2-d1
Files                  → Cloudflare R2       → cloudflare-r2-d1

Monorepo Pattern

单仓多应用模式

apps/web (Next.js)     → Vercel       → vercel-deployment
apps/api (FastAPI)     → Fly.io       → fly-deployment

apps/web (Next.js)     → Vercel       → vercel-deployment
apps/api (FastAPI)     → Fly.io       → fly-deployment

Platform Strengths

平台优势

Vercel

Zero-config Next.js deployment
Preview deployments per PR
Edge functions & middleware
Automatic HTTPS & CDN

Best for: Next.js, React, static sites

零配置Next.js部署
每个PR对应预览部署
边缘函数与中间件
自动HTTPS与CDN

最适合： Next.js、React、静态站点

Fly.io

Run any Docker container
Persistent volumes (SQLite, file storage)
Multiple regions with data locality
WebSocket support

Best for: Python backends, stateful apps, containers

运行任意Docker容器
持久化存储卷（SQLite、文件存储）
多区域部署，支持数据本地化
WebSocket 支持

最适合： Python后端、有状态应用、容器化应用

Cloudflare

Edge-first (runs in 300+ cities)
No cold starts (Workers)
Integrated storage (R2, D1, KV)
Zero egress fees on R2

Best for: Low-latency APIs, edge computing, global scale

边缘优先部署（在300+城市运行）
无冷启动（Workers）
集成存储服务（R2、D1、KV）
R2 无出口流量费用

最适合： 低延迟API、边缘计算、全球规模化应用

Common Deployment Patterns

常见部署模式

Pattern 1: Vercel + Fly.io

模式1：Vercel + Fly.io

User → Vercel (frontend) → Fly.io (API) → SQLite (volume)

Deploy commands:

bash

undefined

User → Vercel (frontend) → Fly.io (API) → SQLite (volume)

部署命令：

bash

undefined

Frontend

vercel --prod

Backend

fly deploy

undefined

fly deploy

undefined

Pattern 2: Full Cloudflare

模式2：全Cloudflare部署

User → Pages (frontend) → Workers (API) → D1 (database) + R2 (files)

Deploy command:

bash

wrangler deploy

User → Pages (frontend) → Workers (API) → D1 (database) + R2 (files)

部署命令：

bash

wrangler deploy

Pattern 3: Hybrid (Vercel + Cloudflare Storage)

模式3：混合部署（Vercel + Cloudflare存储）

User → Vercel (Next.js) → Cloudflare R2 (uploads) + external DB

User → Vercel (Next.js) → Cloudflare R2 (uploads) + external DB

Troubleshooting Quick Links

常见问题排查快速链接

Issue	Check Skill
Vercel deployment issues	`vercel-deployment`
Docker build fails	`fly-deployment`
Volume mount error	`fly-deployment`
D1 query timeout	`cloudflare-r2-d1`
R2 upload fails	`cloudflare-r2-d1`

问题	查看对应技能
Vercel 部署问题	`vercel-deployment`
Docker 构建失败	`fly-deployment`
存储卷挂载错误	`fly-deployment`
D1 查询超时	`cloudflare-r2-d1`
R2 上传失败	`cloudflare-r2-d1`

Cost Comparison (Free Tiers)

免费层成本对比

Platform	Free Tier Highlights
Vercel	100GB bandwidth, serverless functions, preview deploys
Fly.io	3 VMs (256MB), 3GB storage, 160GB bandwidth
Cloudflare	100K Worker requests/day, 10GB R2, 5GB D1

平台	免费层亮点
Vercel	100GB 带宽、无服务器函数、预览部署
Fly.io	3台虚拟机（256MB内存）、3GB存储、160GB带宽
Cloudflare	每日10万次Worker请求、10GB R2存储、5GB D1存储

When to Use Each Skill

各技能适用场景

Use
vercel-deployment
when:

Deploying Next.js or React apps
Configuring vercel.json
Setting environment variables
Troubleshooting build errors

Use
fly-deployment
when:

Writing Dockerfiles for Fly.io
Configuring fly.toml
Setting up volumes or persistent storage
Deploying Python/FastAPI backends

Use
cloudflare-r2-d1
when:

Storing files in R2
Working with D1 SQLite database
Caching with KV
Writing Cloudflare Workers

当你需要以下操作时使用
vercel-deployment
：

部署Next.js或React应用
配置vercel.json
设置环境变量
排查构建错误

当你需要以下操作时使用
fly-deployment
：

为Fly.io编写Dockerfile
配置fly.toml
设置存储卷或持久化存储
部署Python/FastAPI后端

当你需要以下操作时使用
cloudflare-r2-d1
：

在R2中存储文件
操作D1 SQLite数据库
使用KV进行缓存
编写Cloudflare Workers