Question 270244 in Javascript Question posted in
A very good W3school tutorial can be found here.

Javascript – Does MongoDB Node driver create new connection per query?

Check the below code:

const mongodb = require('mongodb');
const express = require('express');

const app = express();

let db;
const options = {};

mongodb.MongoClient.connect('mongodb://localhost:27017/test', options, function(err, database) {
  if (err) throw err;
  db = database.db('test');

  app.listen(process.env.PORT || 3000, function() {
    console.log(`Express.js server up.`);
  });
});

app.get('/', function (req, res) {
  db.collection('test').countDocuments({}, function(err, count) {
    return res.json({'documentCount': count});
  });
});

It is a simple code to do a count operation in MongoDB. My understanding about the above code is that I create a single MongoDB connection and assign it to db, then reuse it where I need to execute queries.

But this Stack Overflow answer states:

Each .connect() creates a new connection pool, not a connection

Also, this article states:

The pool size is the option that determines how many queries may be run using the same connection in parallel, similar to the number of lanes in a tunnel.

If the connect() method is creating a pool instead of connection, at what point are the connections getting created? Does each query create a new connection (or reuse) in the specified pool?

If I execute .connect() only once in my code with poolSize: 5. Does it create 5 connections immediately?

Tags:

2

Answers


  1. 0

    When you call mongodb.MongoClient.connect, you’re essentially creating a connection pool, not just a single connection to MongoDB. This pool contains a number of connections that can be reused for different queries, which is an efficient way to handle database interactions in an environment where there might be many concurrent operations happening.

    A connection pool is a cache of open, ready-to-use database connections maintained by the driver. Your application can grab connections from this pool, use them for database operations, and then return them back to the pool once done​. Source

    In your code, when you call mongodb.MongoClient.connect, it initializes a connection pool, and the db variable holds a reference to this pool. Then, whenever you execute a query like db.collection('test').countDocuments({}, function(err, count) {...}, a connection is borrowed from the pool to execute the query, and returned to the pool once the query execution is complete.

    Login or Signup to reply.
  2. 0

    From the docs, the connect() methods connects the client you created with mongodb.MongoClient to a
    MongoDB Atlas deployment, a MongoDB instance, or a replica set the mongoDB driver, and this creates a connection pool.

    The number of concurrent connections in the pool is then managed internally by the mongoDB driver, as required by the workload.

    However, in the connection options you can override the default settings, including how many concurrent connections there can be for this client (with the maxPoolSize and minPoolSize options; default value is a 100).

    But I think overriding default settings would only be recommended if you are collecting metrics and monitoring MongoDb closely. Then, it may be that you want to fine tune it with the override settings

    Please refer to THIS very good answer in the mongoDb community forums, for more details.

    Login or Signup to reply.
Please signup or login to give your own answer.

Back To Top
Search