DataLoader integration in NestJS

Introduction

DataLoader helps reduce repeated database calls by batching and caching requests. It makes your NestJS app faster and more efficient.

When you have multiple requests for the same data in one operation.

When you want to avoid making many database calls for related data.

When using GraphQL resolvers that fetch nested or repeated data.

When you want to improve performance by caching repeated queries.

When you want to simplify data fetching logic by batching requests.

Syntax

NestJS

import DataLoader from 'dataloader';

const loader = new DataLoader(async (keys) => {
  // batch load function returns results in order of keys
  const results = await batchFetchFromDatabase(keys);
  return keys.map(key => results.find(r => r.id === key));
});

The batch load function receives an array of keys and must return results in the same order.

DataLoader caches results during a request to avoid duplicate fetches.

Examples

This example batches user ID requests and returns users in the same order.

NestJS

const userLoader = new DataLoader(async (userIds) => {
  const users = await userService.findByIds(userIds);
  return userIds.map(id => users.find(user => user.id === id));
});

Similar pattern for loading posts by IDs efficiently.

NestJS

const postLoader = new DataLoader(async (postIds) => {
  const posts = await postService.findByIds(postIds);
  return postIds.map(id => posts.find(post => post.id === id));
});

Sample Program

This example shows a NestJS service using DataLoader to batch user ID requests. It caches during one request scope and returns user names in order.

NestJS

import { Injectable, Scope } from '@nestjs/common';
import DataLoader from 'dataloader';

interface User {
  id: number;
  name: string;
}

@Injectable({ scope: Scope.REQUEST })
export class UserLoader {
  public readonly loader: DataLoader<number, User>;

  constructor(private readonly userService: UserService) {
    this.loader = new DataLoader(async (userIds: readonly number[]) => {
      const users = await this.userService.findByIds(userIds as number[]);
      return userIds.map(id => users.find(user => user.id === id));
    });
  }
}

// Usage in resolver or service
async function getUserNames(userLoader: UserLoader, ids: number[]) {
  const users = await Promise.all(ids.map(id => userLoader.loader.load(id)));
  return users.map(user => user?.name ?? 'Unknown');
}

// Mock UserService for demonstration
class UserService {
  private users = [
    { id: 1, name: 'Alice' },
    { id: 2, name: 'Bob' },
    { id: 3, name: 'Carol' },
  ];

  async findByIds(ids: number[]): Promise<User[]> {
    // Simulate async DB call
    return this.users.filter(user => ids.includes(user.id));
  }
}

// Demo run
(async () => {
  const userService = new UserService();
  const userLoader = new UserLoader(userService);
  const names = await getUserNames(userLoader, [1, 2, 3]);
  console.log(names.join(', '));
})();

Important Notes

Use request-scoped providers in NestJS to create a new DataLoader instance per request.

Always return results in the same order as the keys array to avoid mismatches.

DataLoader caches only during one request; it does not cache across requests.

Summary

DataLoader batches and caches data fetching to improve performance.

Use it in NestJS with request-scoped providers for safe caching.

Always keep the batch function output order matching the input keys.

Practice

1. What is the main purpose of using DataLoader in a NestJS application?

easy

A. To handle HTTP routing and middleware

B. To serve static files efficiently

C. To manage database schema migrations

D. To batch and cache multiple data requests to improve performance

Solution

1. Step 1: Understand DataLoader's role

   DataLoader is designed to batch and cache data fetching operations to reduce redundant calls.

2. Step 2: Compare with other options

   Options A, C, and D relate to other NestJS features, not DataLoader's purpose.

3. Final Answer:

   To batch and cache multiple data requests to improve performance -> Option D

4. Quick Check:

   DataLoader = batch + cache [OK]

Hint: DataLoader batches requests to avoid repeated fetching [OK]

Common Mistakes:

• Confusing DataLoader with routing or middleware
• Thinking DataLoader manages database migrations
• Assuming DataLoader serves static files

2. Which of the following is the correct way to create a DataLoader instance inside a NestJS service?

easy

A. const loader = new DataLoader(batchFunction());

B. const loader = DataLoader.create(batchFunction);

C. const loader = new DataLoader(keys => batchFunction(keys));

D. const loader = DataLoader(batchFunction);

Solution

1. Step 1: Recall DataLoader instantiation syntax

   DataLoader is instantiated with new DataLoader(batchFunction), where batchFunction is a function receiving keys.

2. Step 2: Analyze options

   const loader = new DataLoader(keys => batchFunction(keys)); correctly uses new DataLoader(keys => batchFunction(keys)). Options B, C, and D use incorrect syntax or call batchFunction immediately.

3. Final Answer:

   const loader = new DataLoader(keys => batchFunction(keys)); -> Option C

4. Quick Check:

   Use new DataLoader(batchFunction) [OK]

Hint: Use 'new DataLoader' with a batch function callback [OK]

Common Mistakes:

• Calling batchFunction immediately instead of passing it
• Using DataLoader as a function without 'new'
• Using a non-function argument

3. Given the following batch function in a NestJS DataLoader service:

async function batchUsers(ids: readonly number[]) {
  return ids.map(id => ({ id, name: `User${id}` }));
}

What will be the output of await loader.load(2) if loader is created with this batch function?

medium

A. { id: 2, name: 'User2' }

B. [{ id: 2, name: 'User2' }]

C. undefined

D. Error: batch function must return array of same length

Solution

1. Step 1: Understand batch function output

   The batch function returns an array of user objects matching the input ids in order.

2. Step 2: Understand loader.load behavior

   loader.load(2) returns a single user object corresponding to id 2, not an array.

3. Final Answer:

   { id: 2, name: 'User2' } -> Option A

4. Quick Check:

   load(id) returns single item from batch array [OK]

Hint: load() returns one item, batch returns array [OK]

Common Mistakes:

• Expecting load() to return an array
• Confusing batch function return with load return
• Assuming batch function returns undefined

4. You wrote this batch function for DataLoader:

async function batchPosts(ids: readonly number[]) {
  const posts = await getPostsByIds(ids);
  return posts;
}

But you get wrong results when loading posts. What is the likely problem?

medium

A. The returned array posts is not in the same order as ids

B. The batch function should not be async

C. You must return a Map instead of an array

D. The function should return a Promise of a single post, not an array

Solution

1. Step 1: Check batch function contract

   DataLoader requires the batch function to return results in the same order as the keys array.

2. Step 2: Analyze getPostsByIds behavior

   If getPostsByIds returns posts unordered, the batch function breaks this contract causing wrong results.

3. Final Answer:

   The returned array posts is not in the same order as ids -> Option A

4. Quick Check:

   Batch function must preserve order [OK]

Hint: Batch function output order must match input keys order [OK]

Common Mistakes:

• Ignoring order of returned array
• Thinking batch function can't be async
• Returning Map instead of array

5. In a NestJS GraphQL resolver, you want to integrate DataLoader to batch user requests per HTTP request. Which approach correctly ensures a new DataLoader instance is created for each request to avoid caching issues across users?

hard

A. Create a DataLoader instance once in the module constructor

B. Create a DataLoader instance inside a request-scoped provider injected into the resolver

C. Create a single global DataLoader instance as a singleton service

D. Use a static DataLoader instance shared across all requests

Solution

1. Step 1: Understand DataLoader caching scope

   DataLoader caches results per instance, so sharing one instance globally causes cross-request data leaks.

2. Step 2: Use request-scoped provider

   Creating a DataLoader instance per request via a request-scoped provider ensures isolated caches per user request.

3. Final Answer:

   Create a DataLoader instance inside a request-scoped provider injected into the resolver -> Option B

4. Quick Check:

   Request-scoped DataLoader avoids cross-request cache [OK]

Hint: Use request-scoped provider for per-request DataLoader [OK]

Common Mistakes:

• Using singleton DataLoader causing shared cache
• Creating DataLoader once in module constructor
• Using static instance shared globally