DEV Community

Cover image for Is Your Code Running Slow? The Impact of Neglecting Function Memoization
Jaimal Dullat
Jaimal Dullat

Posted on • Originally published at jaimaldullat.Medium

Is Your Code Running Slow? The Impact of Neglecting Function Memoization

Have you ever wondered why one particular function call in your code seems to run slower than others? Or maybe you’ve noticed some redundancy where the same function gets called multiple times with the same arguments. If this sounds familiar, function memoization could help optimize your code.

In this blog post, we’ll discuss:

  • What is memoization and why it’s useful

  • Implementing basic memoization in JavaScript

  • When not to use memoization

  • Conclusion

1. What is Memoization?

Function memoization is a programming technique that helps speed up repeated function calls by caching previous computation results. Instead of recalculating the output, the function returns previously calculated values. This avoids unnecessary re-computation and improves performance, especially for functions that take a long time to execute or make resource-intensive operations like network or database calls.

For example, say we have a function that calculates the factorial of a number. Calling this function multiple times with the same input would result in duplicate work:

function factorial(n) {
  if (n <= 1) return 1;
  return n * factorial(n - 1);
}

factorial(5); // 120 
factorial(5); // Recalculates 120
Enter fullscreen mode Exit fullscreen mode

By memoizing this function, we can store the results of previous computations in a cache:

const cache = {};

function memoizedFactorial(n) {
  if (n in cache) return cache[n];

  let result = factorialHelper(n);
  cache[n] = result;
  return result;
}

function factorialHelper(n) {
  if (n <= 1) return 1;
  return n * factorialHelper(n - 1);
}

memoizedFactorial(5); // 120
memoizedFactorial(5); // Returns cached 120
Enter fullscreen mode Exit fullscreen mode

Now repeated calls with the same input are lightning fast! Memoization trades storage space for computation time.

2. Basic Memoization in JavaScript

There are a few different approaches to implementing basic memoization in JavaScript.

Using a Cache Object

The simplest way is to add a cache object and check it before each function call:

// Cache
const cache = {};

// Function to be memoized
function add(a, b) {
  console.log(`Adding ${a} and ${b}`);
  return a + b;
}

// Memoize the 'add' function using the 'memoize' higher-order function
const memoizedAdd = memoize(add);

function memoize(fn) {
  return function() {
    const key = JSON.stringify(arguments);

    if (key in cache) {
      console.log(`Cache hit for key: ${key}`);
      return cache[key];
    }

    const result = fn.apply(this, arguments);
    cache[key] = result;
    console.log(`Cache miss for key: ${key}, storing result: ${result}`);
    return result;
  };
}

// Example usage
console.log(memoizedAdd(2, 3)); // This will calculate and cache the result
console.log(memoizedAdd(2, 3)); // This will retrieve the result from the cache

console.log(memoizedAdd(4, 5)); // This will calculate and cache the result
console.log(memoizedAdd(2, 3)); // This will retrieve the result from the cache

console.log(memoizedAdd(6, 7)); // This will calculate and cache the result

console.log(cache); // You can see the cached results in the 'cache' object
Enter fullscreen mode Exit fullscreen mode

In this example, the add function is memoized using the memoize higher-order function. The cache is used to store the results of function calls based on the serialized arguments. When the same arguments are provided again, the memoized function retrieves the result from the cache, reducing the need to recompute it. This helps improve the performance for functions with repetitive or expensive computations.

Using a WeakMap

A WeakMap allows us to cache based on objects instead of strings. This avoids serialization:

const cache = new WeakMap();

function memoize(fn) {
  return function(...args) {
    const cacheKey = args[0];

    if (cache.has(cacheKey)) {
      console.log('Cache hit');
      return cache.get(cacheKey);
    }

    console.log('Cache miss');
    const result = fn.apply(this, args);
    cache.set(cacheKey, result);

    return result;
  };
}

function expensiveCalculation(obj) {
  // Simulating an expensive computation
  let result = 0;
  for (const key in obj) {
    if (obj.hasOwnProperty(key)) {
      result += obj[key];
    }
  }
  return result;
}

const memoizedExpensiveCalculation = memoize(expensiveCalculation);

const obj1 = {
  a: 1,
  b: 2
};
const obj2 = {
  a: 1,
  b: 2
};

const result1 = memoizedExpensiveCalculation(obj1); // Cache miss, expensive calculation
const result2 = memoizedExpensiveCalculation(obj1); // Cache hit, result retrieved from cache

console.log(result1); // Output: Cache miss
console.log(result2); // Output: Cache hit

const result3 = memoizedExpensiveCalculation(obj2); // Cache miss, as it's a different object
console.log(result3); // Output: Cache miss
Enter fullscreen mode Exit fullscreen mode

In this example, we define a memoize function that takes another function fn as an argument and returns a memoized version of that function. The cache is a WeakMap used to store the results of expensive calculations, with the args[0] object as the cache key.

When we call memoizedExpensiveCalculation(obj1) for the first time, it performs the expensive calculation and stores the result in the cache. When we call it with the same object obj1 again, it detects a cache hit and returns the cached result without re-computing. The WeakMap automatically cleans up the cache if the object is no longer referenced, making it suitable for caching based on objects.

Using Memoization with Async Functions

Fetching data from external APIs or databases often takes considerable time. We can optimize async operations with memoization:

// Cache
const cache = new Map();

function memoizedFetch(url) {
  if (cache.has(url)) {
    return cache.get(url);
  }

  const promise = fetch(url)
    .then((res) => res.json())
    .then((data) => {
      cache.set(url, data); // Cache the JSON response
      return data;
    });

  cache.set(url, promise);

  return promise;
}

// Example usage
async function fetchDataWithMemoization(url) {
  try {
    const data = await memoizedFetch(url);
    console.log("Data from cache or API:", data);
  } catch (error) {
    console.error("Error fetching data:", error);
  }
}

// First call, fetches data from the API and caches it
fetchDataWithMemoization("https://api.example.com/data");

// Second call with the same URL, retrieves data from the cache
fetchDataWithMemoization("https://api.example.com/data");

// Third call with a different URL, fetches data from the API and caches it
fetchDataWithMemoization("https://api.example.com/another-data");
Enter fullscreen mode Exit fullscreen mode

In this example, the memoizedFetch function caches the promises and their results, ensuring that if you request the same URL multiple times, it will return the cached promise for subsequent calls. This can significantly optimize your code by preventing redundant API requests.

There are some advanced memoization techniques to optimize for different use cases.

When Not to Memoize

Memoization is a useful technique for optimizing function calls by caching their results. However, it’s not always the right choice. Here are some situations when you might want to avoid memoization:

  1. Freshness of Data — If the data you’re working with needs to be up-to-date all the time, memoization might not be appropriate. Memoization is based on the assumption that the same input will produce the same output, so if the data changes frequently, you could end up with stale or outdated results.

  2. Memory Constraints — Caching results can consume memory, especially if you have a large number of unique inputs. If your application has strict memory constraints, excessive memoization might not be feasible.

  3. Dynamic Inputs — When function inputs are highly dynamic and you can’t predict which inputs will be used, or when the number of unique inputs is very large, it can be challenging to manage the cache efficiently.

  4. Stateful Operations — Memoization is not suitable for stateful or side-effectful operations. If a function has side effects or relies on external state that can change, memoization might lead to unexpected behavior.

  5. Recursive Functions — Be cautious when memoizing recursive functions. While it can optimize some recursive algorithms, it might lead to stack overflow errors if used inappropriately.

  6. Highly Variable Inputs — If the function’s inputs vary widely and unpredictably, memoization might not provide significant performance benefits. It’s most effective when there’s a high probability of the same inputs being reused.

  7. Memory Management Overhead — Managing the cache itself can introduce overhead. In some cases, the time and resources spent on managing the cache could outweigh the performance gains from memoization.

  8. Complex Cache Invalidation — Implementing cache invalidation logic can be challenging, especially in cases where the inputs or data change frequently. Maintaining a valid cache can become complex and error-prone.

  9. Multi-threading and Concurrency — In multi-threaded or concurrent environments, you need to be careful about race conditions when using memoization. Careful synchronization is required to ensure the cache is accessed safely.

  10. Premature Optimization — Don’t introduce memoization prematurely. It’s best applied when you’ve identified a performance bottleneck in your application and have measured that memoization would provide a noticeable improvement.

Conclusion

Memoization is a powerful technique, but it’s not a one-size-fits-all solution. It should be used thoughtfully, considering the nature of the problem and the specific use case. Careful consideration of factors like data freshness, memory constraints, and the nature of the inputs is essential to determine whether memoization is beneficial or not.

🔥 Wait! 🔥

Give that like button some love! And if you’re feeling extra cheeky, hit follow too!

Follow me on Instagram: Click Here

Top comments (0)