5 Optimization Techniques to Make Turborepo Builds 10x

🎯 What You’ll Learn (Reading Time: 10 minutes)

TL;DR: Learn 5 optimization techniques to reduce Turborepo build time from 10 minutes to 1 minute. Includes practical benchmarks from caching strategies to Remote Cache. What You’ll Learn:

• ✅ Master Turborepo caching strategies
• ✅ Reduce team build time by 90% with Remote Cache
• ✅ Master parallel execution optimization techniques
• ✅ Improve build speed through dependency optimization Real-World Applications:
• 🔥 Reduce CI/CD build time by 10x
• 🔥 Maximize team development productivity
• 🔥 Cut build server costs by 90%
• 🔥 Minimize developer wait time Difficulty: Intermediate | Performance Gain: 10x | Benchmarks: Included

---

Problem: “Why is Turborepo so slow?”

If your builds are slow despite using Turborepo, you likely haven’t optimized it properly. Real Case Study:

Before optimization: 10-minute full build
After optimization: 1-minute full build
━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Improvement: 10x faster ⚡

This guide covers 5 battle-tested optimization techniques.

Optimization 1: Perfect Cache Configuration

Problem: Missing outputs Configuration

The most common mistake. Without proper outputs configuration, caching won’t work.

// ❌ Wrong configuration
{
  "pipeline": {
    "build": {
      "outputs": []  // No caching!
    }
  }
}
// ✅ Correct configuration
{
  "pipeline": {
    "build": {
      "outputs": [
        ".next/**",
        "!.next/cache/**",
        "dist/**",
        "build/**"
      ]
    }
  }
}

Checklist:

{
  "pipeline": {
    "build": {
      "outputs": [
        ".next/**",           // Next.js
        "dist/**",            // General builds
        "build/**",           // CRA, Vite
        ".nuxt/**",           // Nuxt
        "out/**",             // Next.js export
        "public/build/**"     // Remix
      ]
    },
    "test": {
      "outputs": [
        "coverage/**"         // Test coverage
      ]
    }
  }
}

Impact:

No cache: 10 minutes every time
With cache: 10 minutes first build, 10 seconds after
━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Improvement: 60x faster 🚀

---

Optimization 2: Dependency Graph Optimization

Problem: Unnecessary Dependencies

// ❌ Everything depends on everything
{
  "pipeline": {
    "build": {
      "dependsOn": ["^build", "lint", "test"]  // Too many!
    }
  }
}
// ✅ Only necessary dependencies
{
  "pipeline": {
    "build": {
      "dependsOn": ["^build"]  // Only upstream builds
    },
    "deploy": {
      "dependsOn": ["build", "test"]  // Test only on deploy
    }
  }
}

Dependency Minimization Strategy:

{
  "pipeline": {
    // Build: only upstream dependencies
    "build": {
      "dependsOn": ["^build"]
    },
    
    // Lint: independent execution
    "lint": {
      "dependsOn": []
    },
    
    // Test: run after build
    "test": {
      "dependsOn": ["build"]
    },
    
    // Type check: only upstream builds
    "type-check": {
      "dependsOn": ["^build"]
    }
  }
}

Impact:

Excessive dependencies: sequential execution (slow)
Minimal dependencies: parallel execution (fast)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Improvement: 3-5x faster

---

Optimization 3: Leverage Remote Cache

Vercel Remote Cache (Free)

# 1. Connect Vercel account
npx turbo login
# 2. Link project
npx turbo link
# 3. Build (automatically uses Remote Cache)
turbo build

Impact:

Scenario: Developer A builds → Developer B builds
Without Remote Cache:
- A: 10 minutes
- B: 10 minutes
- Total: 20 minutes
With Remote Cache:
- A: 10 minutes (create cache)
- B: 10 seconds (download cache)
- Total: 10 minutes 10 seconds
━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Team-wide time savings: 50%

Self-Hosted Remote Cache Server

# Simple setup with Docker
docker run -d -p 8080:8080 \
  -v turborepo-cache:/cache \
  ghcr.io/fox1t/turborepo-remote-cache
# Configure in turbo.json
{
  "remoteCache": {
    "url": "http://localhost:8080"
  }
}

---

Optimization 4: Maximize Parallel Execution

Problem: Sequential Execution

// ❌ Sequential execution (slow)
{
  "pipeline": {
    "build": {
      "dependsOn": ["lint", "test", "type-check"]
    }
  }
}

lint (2min) → test (3min) → type-check (1min) → build (4min)
Total: 10 minutes

Solution: Parallel Execution

// ✅ Parallel execution (fast)
{
  "pipeline": {
    "build": {
      "dependsOn": ["^build"]
    },
    "lint": {},
    "test": {},
    "type-check": {}
  }
}

lint (2min) ┐
test (3min) ├─ parallel execution → build (4min)
type-check (1min) ┘
Total: 7 minutes (3 minutes saved)

Parallel Execution Commands:

# Run multiple tasks simultaneously
turbo run lint test type-check --parallel
# Build all packages simultaneously
turbo run build --parallel
# Limit concurrency (prevent CPU overload)
turbo run build --concurrency=4

---

Optimization 5: Selective Builds (Filtering)

Build Only Changed Packages

# Build only changed packages
turbo run build --filter=...[HEAD^1]
# Build specific package with dependencies
turbo run build --filter=web...
# Build specific package only (exclude dependencies)
turbo run build --filter=web
# Select multiple packages
turbo run build --filter=web --filter=docs
# Only when specific path changes
turbo run build --filter=./apps/*

Practical Examples:

# PR builds: only changed
turbo run build test --filter=...[origin/main]
# Specific app deployment: only that app
turbo run build --filter=web...
# Package change: only affected apps
turbo run build --filter=...@ui

Impact:

Full build: 10 minutes
Filtered build: 2 minutes
━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Improvement: 5x faster

---

Comprehensive Optimization Strategy

Before: Pre-Optimization

// turbo.json (before optimization)
{
  "pipeline": {
    "build": {
      "outputs": [],  // ❌ No caching
      "dependsOn": ["lint", "test"]  // ❌ Sequential execution
    }
  }
}

# Build command
turbo run build
# Result:
# lint: 2 minutes
# test: 3 minutes
# build: 5 minutes
# Total: 10 minutes (no cache)

After: Post-Optimization

// turbo.json (after optimization)
{
  "pipeline": {
    "build": {
      "outputs": [".next/**", "dist/**"],  // ✅ Cache enabled
      "dependsOn": ["^build"]  // ✅ Minimal dependencies
    },
    "lint": {},  // ✅ Parallel execution
    "test": {}   // ✅ Parallel execution
  }
}

# Build command
turbo run build lint test --parallel
# Result:
# lint, test, build run in parallel: 5 minutes
# Cache hit: 10 seconds
# Total: 5 minutes first build, 10 seconds after

Improvement:

Before optimization: 10 minutes
After optimization: 10 seconds (cache hit)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Improvement: 60x faster ⚡⚡⚡

---

Real-World Benchmarks

Based on actual project (10 packages, Next.js + React):

Optimization Stage	Build Time	Improvement
Default config	10 minutes	-
+ outputs config	10min → 2min (cache hit)	5x
+ dependency optimization	10min → 6min	1.7x
+ parallel execution	6min → 4min	1.5x
+ Remote Cache	4min → 30sec (team cache)	8x
+ filtering	4min → 1min (changed only)	4x
Total	10min → 30sec	20x

---

Checklist

If your builds are slow, check these: Caching:

• Set build output paths in outputs?
• Is .turbo folder in .gitignore?
• Connected Remote Cache? (team projects) Dependencies:
• Any unnecessary tasks in dependsOn?
• No circular dependencies?
• Simplified to depend only on ^build? Parallelization:
• Running independent tasks in parallel?
• Using --parallel flag?
• Is --concurrency setting appropriate? Filtering:
• Building only changed parts in CI?
• Filtering when deploying specific apps?

---

Ready-to-Use Configuration

Optimized turbo.json

{
  "$schema": "https://turbo.build/schema.json",
  "globalDependencies": ["**/.env.*local"],
  "pipeline": {
    "build": {
      "dependsOn": ["^build"],
      "outputs": [
        ".next/**",
        "!.next/cache/**",
        "dist/**",
        "build/**"
      ],
      "env": [
        "NEXT_PUBLIC_*",
        "VITE_*"
      ]
    },
    "dev": {
      "cache": false,
      "persistent": true
    },
    "lint": {
      "outputs": []
    },
    "test": {
      "outputs": ["coverage/**"],
      "dependsOn": []
    },
    "type-check": {
      "dependsOn": ["^build"],
      "outputs": []
    }
  }
}

package.json Scripts

{
  "scripts": {
    "build": "turbo run build",
    "build:changed": "turbo run build --filter=...[HEAD^1]",
    "dev": "turbo run dev --parallel",
    "lint": "turbo run lint --parallel",
    "test": "turbo run test --parallel",
    "clean": "turbo run clean && rm -rf node_modules",
    "clean:cache": "rm -rf .turbo"
  }
}

CI/CD Optimization (GitHub Actions)

name: CI
on: [push, pull_request]
jobs:
  build:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v3
      
      # pnpm cache
      - uses: pnpm/action-setup@v2
        with:
          version: 8
      
      - uses: actions/setup-node@v3
        with:
          node-version: 18
          cache: 'pnpm'
      
      - run: pnpm install --frozen-lockfile
      
      # Turborepo Remote Cache
      - name: Build
        run: pnpm turbo run build --filter=...[HEAD^1]
        env:
          TURBO_TOKEN: ${{ secrets.TURBO_TOKEN }}
          TURBO_TEAM: ${{ secrets.TURBO_TEAM }}
      
      # Test only changed
      - name: Test
        run: pnpm turbo run test --filter=...[HEAD^1]

CI Build Time:

Before optimization: 15 minutes
After optimization: 2-3 minutes (cache + filtering)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Improvement: 5-7x faster

---

Advanced Optimization

1. Leverage Incremental Builds

{
  "pipeline": {
    "build": {
      "dependsOn": ["^build"],
      "outputs": [".next/**"],
      "cache": true,
      "inputs": [
        "src/**",
        "public/**",
        "package.json",
        "next.config.js",
        "tsconfig.json"
      ]
    }
  }
}

Specifying inputs triggers rebuild only when those files change.

2. Environment Variable Cache Keys

{
  "pipeline": {
    "build": {
      "env": [
        "NEXT_PUBLIC_API_URL",
        "NEXT_PUBLIC_ENV"
      ]
    }
  }
}

Cache is invalidated when environment variables change.

3. Selective Caching

{
  "pipeline": {
    "dev": {
      "cache": false,  // No cache in dev mode
      "persistent": true
    },
    "build": {
      "cache": true  // Cache only builds
    }
  }
}

---

Performance Monitoring

Build Analysis

# Detailed logs
turbo run build --verbosity=2
# Timing information
turbo run build --profile=profile.json
# Check cache status
turbo run build --dry-run

Find Bottlenecks

# Check build time per package
turbo run build --graph
# Visualize dependency graph
turbo run build --graph=graph.html

---

Practical Tips

1. Local Development Optimization

# Build only changed app
turbo run build --filter=web...
# Parallel dev servers
turbo run dev --parallel --filter=web --filter=api

2. CI Optimization

# PR: only changed
turbo run build test --filter=...[origin/main]
# main: full build + cache refresh
turbo run build test

3. Deployment Optimization

# Build and deploy specific app only
turbo run build --filter=web...
cd apps/web && vercel deploy

---

Summary

5 Optimization Techniques

1. Cache Configuration: Perfect outputs setup → 60x improvement
2. Dependency Optimization: Remove unnecessary dependsOn → 3-5x improvement
3. Remote Cache: Share cache across team → 50% team-wide savings
4. Parallel Execution: --parallel flag → 2-3x improvement
5. Filtering: Build only changed → 5x improvement

Overall Impact

Before optimization: 10 minutes
After optimization: 30 seconds (cache hit)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Improvement: 20x faster 🚀🚀🚀

Immediate Action Checklist

Today:

• Set outputs in turbo.json
• Remove unnecessary dependsOn
• Add --parallel flag This Week:
• Connect Remote Cache
• Apply filtering in CI
• Measure and compare build times This Month:
• Roll out Remote Cache team-wide
• Optimize dependency graph
• Build monitoring dashboard

---

Learn More

• Complete Turborepo Guide - From basics to advanced
• pnpm Workspace Guide - Package manager optimization
• Monorepo Architecture Design - Structure design best practices Dramatically reduce your build time with Turborepo! 🚀

---

Frequently Asked Questions (FAQ)

Q. When would I use this in practice?

A. A practical optimization guide to reduce Turborepo build time from 10 minutes to 1 minute.

Q. What should I read before this?

A. Follow the previous article or related articles links at the bottom of each post to learn in sequence.

Q. Where can I study this more deeply?

A. Check cppreference and the relevant library’s official documentation. The reference links at the end of the article are also worth using.

---

Related Articles (Internal Links)

Other articles related to this topic.

• Turborepo 완벽 가이드 | Monorepo·빌드 캐싱·병렬 실행·pnpm·실전 활용
• pnpm Complete Guide
• Docker Compose Complete Guide | Multi-Container Apps

---

Keywords Covered in This Article (Related Search Terms)

This article covers Turborepo, Monorepo, Performance, Build, Optimization, Cache.