const { performance } = require("perf_hooks");
```


Now let's write a function that calculates the Fibonacci sequence of nth number recursively.



```javascript
function fibonacci(n) {
  if (n === 0 || n === 1)
    return n;
  else
    return fibonacci(n - 1) + fibonacci(n - 2);
}
```



This function performs well for small values of “n”. However, performance quickly degrades as “n” increases. This is because the two recursive calls repeat the same work. For example, to compute the 50th Fibonacci number, the recursive function must be called over 40 billion times (40,730,022,147 times to be specific)! We'll see this visually later.

## A memoized Fibonacci function.
In the memoized version of the Fibonacci function When f() is returned, its closure allows it to continue to access the “memo” object, which stores all of its previous results. Each time f() is executed, it first checks to see if a result exists for the current value of “n”. If it does, then the cached value is returned. Otherwise, the original Fibonacci code is executed. Note that “memo” is defined outside of f() so that it can retain its value over multiple function calls.



```javascript
var memoizeFibonacci = function() {
  var memo = {};

  function f(n) {
    var value;

    if (n in memo) {
      value = memo[n];
    } else {
      if (n === 0 || n === 1)
        value = n;
      else
        value = f(n - 1) + f(n - 2);

      memo[n] = value;
    }

    return value;
  }

  return f;
};

```


## Comparing performance with `perf-hooks`. 

Let's visualize the time it takes to compute 30th Fibonacci number with both functions.



```javascript
//un-optimized
// time before function is executed
const startTime = performance.now();
fibonacci(20);
// time after function has completed computation
const endTime = performance.now();

console.log("Un-optimized time", endTime - startTime);

// memoized
const startTime2 = performance.now();
memoizeFibonacci(20);
// time after function has completed computation
const endTime2 = performance.now();

console.log("Optimized time", endTime2 - startTime2);
```





```javascript
//result

Un-optimized:  1020.0609370004386
Optimized:  0.049122998490929604
```



You can see we already increased the time to compute by a magnitude of over 20000. That's just for a sequence of 30 numbers. This example is quite simple and may not look similar to your everyday tasks, but if you looked deeply there are a couple of things that can be optimized in your program. Keep in mind that memoization is just one optimization method, there are countless different strategies. Don't be the hammer guy that treats every problem like a nail.

Note also that we have barely scratched the surface of memoization, this is just to open our minds to the possibilities.

The fact that it works does not mean it can't be improved. Go forth and optimize!

PS: The title is a bit of an exaggeration. It just happened to be the 97th title that crossed my mind😃