Salts in Key Derivation
What You'll Learn
You'll learn how salts prevent attackers from pre-computing massive tables of passwords and ensure that identical passwords produce different encryption keys.
The Problem: Rainbow Tables
Imagine two users choose the same password: "Summer2024!". Without salts, running that password through a KDF (like PBKDF2 or Argon2) would produce identical derived keys for both users. Attackers exploit this by pre-computing rainbow tables—massive databases mapping common passwords to their derived keys. They compute billions of passwords once, then instantly crack any matching hash.
How Salts Solve This
A salt is a unique, random value added to each password before derivation. Even if two users have identical passwords, different salts ensure completely different derived keys:
User A: DeriveKey("Summer2024!" + salt_A) → key_A
User B: DeriveKey("Summer2024!" + salt_B) → key_B
The keys are totally different, even though the passwords match.
Salt Properties
- Random: Generated using a cryptographically secure random number generator
- Unique: Different for every password (or every user/system)
- Non-secret: Salts can be stored alongside the derived key—they're not passwords themselves
- Sufficient length: Typically 128 bits (16 bytes) minimum
Why Salts Work
Salts force attackers to compute rainbow tables separately for every single salt value. With a 128-bit salt, that's 2^128 possible tables—computationally infeasible. Attackers must resort to attacking passwords one-by-one, which slow KDFs like Argon2 make prohibitively expensive.
Analogy
Think of salts like adding a random spice blend to each dish. Even if two chefs use the same base recipe (password), the unique spice mix (salt) makes each dish taste completely different.
Key Takeaway: Salts are random values added to passwords during key derivation that prevent rainbow table attacks and ensure identical passwords produce unique keys—they must be unique, sufficiently long, and cryptographically random.