Why O(1) Beats O(n): Understanding Big O Through Simple Box Examples

Introduction

Algorithms are methods for solving problems, and a problem often has many possible algorithms. To decide which algorithm is better or more efficient, we use Big O notation, which includes:

O(n) – linear time operation

O(1) – constant time operation

O(log n) – logarithmic time operation

For example, Redis documentation provides a complexity description for each command.

Understanding the purpose of Big O helps us improve program efficiency; below are the specific meanings.

O(n) Linear Time Operation

Imagine a box containing many numbered cards (e.g., 1, 2, 3, … 16). We are asked to find the card with the number 6.

We take out one card at a time, check if it is 6, and stop when we find it; otherwise we continue to the next card. In the worst case, all cards are examined.

This method is a linear operation, denoted O(n).

O(1) Constant Time Operation

Suppose a box contains numbers (1, 2, 3, … 16) and outside the box there is a label stating that the box holds 16 numbers.

When someone asks how many numbers are in the box, we can look at the label and instantly answer 16.

This is a constant‑time operation, denoted O(1).

O(log n) Logarithmic Time Operation

Assume a box contains ordered numbers (1, 2, 3, … 16). To find the number 16, we can split the ordered set into two halves, keep the half that may contain 16, and repeat the process. After four steps we locate 16.

16 = 2⁴.

If there are 64 numbers, finding 64 requires 6 steps.

This is a logarithmic operation, denoted O(log n).

Summary

We can see that O(1) is the most powerful—no matter how large the data set, it completes instantly. O(n) performs worst with large data, while O(log n) scales logarithmically, offering a significant speed advantage.

Knowing the meaning of Big O lets us choose better algorithms; for example, the Redis keys command has a complexity of O(n), so it should be used with caution.