Prisma setup in NestJS - Performance & Optimization

Performance: Prisma setup in NestJS

MEDIUM IMPACT

This affects server response time and API data fetching speed, impacting how fast the frontend receives data.

Setting up Prisma client in NestJS for database access

NestJS

import { PrismaClient } from '@prisma/client';
import { Injectable, OnModuleDestroy, OnModuleInit } from '@nestjs/common';

@Injectable()
export class PrismaService extends PrismaClient implements OnModuleInit, OnModuleDestroy {
  async onModuleInit() {
    await this.$connect();
  }

  async onModuleDestroy() {
    await this.$disconnect();
  }
}

@Injectable()
export class UserService {
  constructor(private prisma: PrismaService) {}

  async getUsers() {
    return await this.prisma.user.findMany();
  }
}

Using a singleton PrismaService injected via NestJS DI avoids multiple connections and manages lifecycle properly.

📈 Performance GainSingle DB connection reused, reducing latency and memory overhead

Setting up Prisma client in NestJS for database access

NestJS

import { Injectable } from '@nestjs/common';
import { PrismaClient } from '@prisma/client';

@Injectable()
export class UserService {
  private prisma = new PrismaClient();

  async getUsers() {
    return await this.prisma.user.findMany();
  }
}

Creating a new PrismaClient instance directly in the service causes multiple connections and higher memory use.

📉 Performance CostAdds multiple DB connections, increasing latency and memory usage per request

Performance Comparison

Pattern	DB Connections	Memory Usage	Response Latency	Verdict
Direct PrismaClient in service	Multiple per service instance	High	Higher due to connection overhead	[X] Bad
Singleton PrismaService with DI	Single shared connection	Low	Lower due to reuse	[OK] Good

Rendering Pipeline

Prisma setup affects backend API response time, which influences frontend rendering speed indirectly by controlling data availability.

→Server Processing

→Network Transfer

→Frontend Rendering

⚠️ BottleneckServer Processing (database query execution and connection management)

Core Web Vital Affected

INP

This affects server response time and API data fetching speed, impacting how fast the frontend receives data.

Optimization Tips

1Always use a singleton PrismaService injected via NestJS DI to avoid multiple DB connections.

2Connect and disconnect PrismaClient in lifecycle hooks to manage resources efficiently.

3Monitor API response times in DevTools Network tab to catch Prisma-related delays early.

Performance Quiz - 3 Questions

What is the main performance issue with creating a new PrismaClient instance inside each NestJS service?

A It blocks CSS from loading

B It causes frontend rendering delays directly

C It creates multiple database connections increasing latency and memory use

D It increases bundle size significantly

DevTools: Network

How to check: Open DevTools Network tab, make API requests, and observe response times for data fetching endpoints.

What to look for: Look for consistent and low response times indicating efficient backend Prisma setup.