This Fibonacci example is purposely inefficient to demonstrate an important consideration for your applications. What happens to the Node.js event loop when running long computations? To see the effect, open two browser windows, each opened to the Fibonacci page. In one, enter the number 55 or greater, and in the other, enter 10. Note that the second window freezes, and if you leave it running long enough, the answer will eventually pop up in both windows. What's happening is the Node.js event loop is blocked from processing events because the Fibonacci algorithm is running and does not ever yield to the event loop.

Since Node.js has a single execution thread, processing requests depend on request handlers quickly returning to the event loop. Normally, the asynchronous coding style ensures that the event loop executes regularly.

This is true even for requests that load data from a server halfway around the globe, because the asynchronous I/O is non-blocking and control is quickly returned to the event loop. The naïve Fibonacci function we chose doesn't fit into this model because it's a long-running blocking operation. This type of event handler prevents the system from processing requests and stops Node.js from doing what it's meant to do, namely to be a blisteringly fast web server.

In this case, the long-response-time problem is obvious. Response time quickly escalates to the point where you can take a vacation to Tibet and perhaps get reincarnated as a llama in Peru during the time it takes to respond with the Fibonacci number!

To see this more clearly, create a file named fibotimes.js containing the following code:

const math = require('./math'); 
const util = require('util'); 
 
for (var num = 1; num < 80; num++) {
    let now = new Date().toISOString();
    console.log(`${now} Fibonacci for ${num} = ${math.fibonacci(num)}`);
} 

Now run it. You will get the following output:

$ node fibotimes.js 
2017-12-10T23:04:42.342Z Fibonacci for 1 = 1
2017-12-10T23:04:42.345Z Fibonacci for 2 = 1
2017-12-10T23:04:42.345Z Fibonacci for 3 = 2
2017-12-10T23:04:42.345Z Fibonacci for 4 = 3
2017-12-10T23:04:42.345Z Fibonacci for 5 = 5
...
2017-12-10T23:04:42.345Z Fibonacci for 10 = 55
2017-12-10T23:04:42.345Z Fibonacci for 11 = 89
2017-12-10T23:04:42.345Z Fibonacci for 12 = 144
2017-12-10T23:04:42.345Z Fibonacci for 13 = 233
2017-12-10T23:04:42.345Z Fibonacci for 14 = 377
...
2017-12-10T23:04:44.072Z Fibonacci for 40 = 102334155
2017-12-10T23:04:45.118Z Fibonacci for 41 = 165580141
2017-12-10T23:04:46.855Z Fibonacci for 42 = 267914296
2017-12-10T23:04:49.723Z Fibonacci for 43 = 433494437
2017-12-10T23:04:54.218Z Fibonacci for 44 = 701408733
...
2017-12-10T23:06:07.531Z Fibonacci for 48 = 4807526976
2017-12-10T23:07:08.056Z Fibonacci for 49 = 7778742049
^C

This quickly calculates the first 40 or so members of the Fibonacci sequence, but after the 40th member, it starts taking a couple of seconds per result and quickly degrades from there. It is untenable to execute code of this sort on a single-threaded system that relies on a quick return to the event loop. A web service containing such code would give poor performance to the users.

There are two general ways to solve this problem in Node.js:

• Algorithmic refactoring: Perhaps, like the Fibonacci function we chose, one of your algorithms is suboptimal and can be rewritten to be faster. Or, if not faster, it can be split into callbacks dispatched through the event loop. We'll look at one such method in a moment.
• Creating a backend service: Can you imagine a backend server dedicated to calculating Fibonacci numbers? Okay, maybe not, but it's quite common to implement backend servers to offload work from frontend servers, and we will implement a backend Fibonacci server at the end of this chapter.