DEV Community

loading...
Cover image for Caching in Node.js using Memcached

Caching in Node.js using Memcached

Francisco Mendes
I don't know about you but I'm just getting started. 🍩
・5 min read

I've already written articles about caching using Redis and also explained how we can cache our Api using node-cache.

In each of these articles, I've provided little background on using each of them, and in my opinion, Memcached is something I should have added to the list.

One of the great advantages of using Memcached in your applications is stability and performance, not to mention that the system resources it consumes and the space it takes up is minimal.

As with the examples in the articles mentioned above, I'll do something similar today, which is simple but can be easily replicated in your projects.

Let's code

In today's example I'm going to use my favorite framework, but the client we're going to use is agnostic, that is, the code that is present in this article can be reused for other frameworks.

The framework we are going to use today is tinyhttp which is very similar to Express. The reason for its use to me is pretty obvious, but I recommend visiting the github repository.

In addition, we will still install milliparsec, which is a super lightweight body parser, and the Memcached client we will be using will be memjs.

But today's topic isn't about frameworks, so let's start by installing the following dependencies:

npm i @tinyhttp/app @tinyhttp/logger milliparsec memjs
Enter fullscreen mode Exit fullscreen mode

First we will import our tinnyhttp dependencies and we will register the respective middlewares:

import { App } from "@tinyhttp/app";
import { logger } from "@tinyhttp/logger";
import { json } from "milliparsec";

const app = new App();

app.use(logger());
app.use(json());

// More stuff comes here.

app.listen(3333);
Enter fullscreen mode Exit fullscreen mode

Now we can create our route, which will only hold one parameter, which in this case will be the id:

app.post("/:id", (req, res) => {
  // Logic goes here.
});
Enter fullscreen mode Exit fullscreen mode

First, let's get the id value of the parameters. Next, we will create an object, in which we will have a property with the value of the id and the remaining properties will be all those coming from the http request body.

app.post("/:id", (req, res) => {
  const { id } = req.params;
  const data = { id, ...req.body };
  // More logic goes here.
});
Enter fullscreen mode Exit fullscreen mode

Then we will return a response, which will have status code 201 (to indicate that the data was added to Memcached) and the respective object that was created by us.

app.post("/:id", (req, res) => {
  const { id } = req.params;
  const data = { id, ...req.body };
  return res.status(201).json(data);
});
Enter fullscreen mode Exit fullscreen mode

However, we can't add anything to Memcached yet because it still needs to be configured. So we can already create our client. Like this:

import { App } from "@tinyhttp/app";
import { logger } from "@tinyhttp/logger";
import { json } from "milliparsec";
import { Client } from "memjs";

const app = new App();
const memcached = Client.create();

// Hidden for simplicity
Enter fullscreen mode Exit fullscreen mode

Now we can go back to our route and add Memcached, for that we'll use the .set() method to input some data.

In this method we will pass three arguments, the first one will be our key, which in this case is the id.

The second argument will be the value of that same key, which we must convert to a string.

The third will be the time you want to persist that same data, in seconds.

In addition to this we will have to make our function asynchronous because the .set() method returns a Promise.

app.post("/:id", async (req, res) => {
  const { id } = req.params;
  const data = { id, ...req.body };
  await memcached.set(id, JSON.stringify(data), { expires: 12 });
  return res.status(201).json(data);
});
Enter fullscreen mode Exit fullscreen mode

The next time you access the route, it will persist in Memcached, but we're not there yet.

That's because we still have to create a middleware that checks if there is a key with id equal to what we are passing in the parameters.

If there is a key equal to the id we passed in the parameters, we'll want to return the value of that key so we don't have to access our controller. If it doesn't exist, we go to our controller to create a new key.

If you're confused, relax because it will soon make sense. In this case, let's create a middleware called verifyCache:

const verifyCache = (req, res, next) => {
  // Logic goes here.
};
Enter fullscreen mode Exit fullscreen mode

First let's get the id value that is passed in the parameters.

const verifyCache = (req, res, next) => {
  const { id } = req.params;
  // More logic goes here.
};
Enter fullscreen mode Exit fullscreen mode

Next, we'll use the Memcached client's .get() method. Let's pass two arguments in this method, the first argument will be the id. The second argument will be a callback and will also have two arguments. The first will be the error, the second will be the key value.

const verifyCache = (req, res, next) => {
  const { id } = req.params;
  memcached.get(id, (err, val) => {
    // Even more logic goes here.
  });
};
Enter fullscreen mode Exit fullscreen mode

If an error occurs, we have to handle it as follows:

const verifyCache = (req, res, next) => {
  const { id } = req.params;
  memcached.get(id, (err, val) => {
    if (err) throw err;
    // Even more logic goes here.
  });
};
Enter fullscreen mode Exit fullscreen mode

Now, see that the key value is non-null, we want to return its value, for that we will send a response with status code 200 (to show that it was obtained from Memcached successfully) and we will send our json object (but first it must be converted from string to json).

If the key value is null, we will proceed to the controller.

const verifyCache = (req, res, next) => {
  const { id } = req.params;
  memcached.get(id, (err, val) => {
    if (err) throw err;
    if (val !== null) {
      return res.status(200).json(JSON.parse(val));
    } else {
      return next();
    }
  });
};
Enter fullscreen mode Exit fullscreen mode

Now with the created middleware we just add it to our route:

app.post("/:id", verifyCache, async (req, res) => {
  const { id } = req.params;
  const data = { id, ...req.body };
  await memcached.set(id, JSON.stringify(data), { expires: 12 });
  return res.status(201).json(data);
});
Enter fullscreen mode Exit fullscreen mode

Your final code should look like the following:

import { App } from "@tinyhttp/app";
import { logger } from "@tinyhttp/logger";
import { json } from "milliparsec";
import { Client } from "memjs";

const app = new App();
const memcached = Client.create();

app.use(logger());
app.use(json());

const verifyCache = (req, res, next) => {
  const { id } = req.params;
  memcached.get(id, (err, val) => {
    if (err) throw err;
    if (val !== null) {
      return res.status(200).json(JSON.parse(val));
    } else {
      return next();
    }
  });
};

app.post("/:id", verifyCache, async (req, res) => {
  const { id } = req.params;
  const data = { id, ...req.body };
  await memcached.set(id, JSON.stringify(data), { expires: 12 });
  return res.status(201).json(data);
});

app.listen(3333);
Enter fullscreen mode Exit fullscreen mode

Conclusion

As always, I hope I was brief in explaining things and that I didn't confuse you. Have a great day! 🙌 🥳

Discussion (0)