Seven ways for thousands of strangers to agree on a single truth — compared across the trilemma of speed, security and decentralisation, the chains that run each one, the languages built on top, and where they sit in the stack.
Popularised by Vitalik Buterin: a blockchain struggles to maximise scalability, security and decentralisation at the same time. Every consensus algorithm is a different answer to which corner gets bent.
Real blockchains and their currencies, mapped to the consensus mechanism they run — and why that mechanism fits their design goals.
What developers actually write, tied back to the chains — and through them, to the consensus protocol securing every contract call.
The foundation. A Layer 1 operates its own validator network, its own consensus algorithm and its own security budget. Everything in this explorer with a named algorithm is a Layer 1.
A fast lane built on top of an L1. It executes transactions off the base chain, then posts proofs or checkpoints back down — inheriting the L1's consensus security instead of running a full one of its own.
Why it matters here: consensus lives at Layer 1. Rollups like Arbitrum don't run a full consensus algorithm of their own — they settle on Ethereum and inherit its Proof-of-Stake finality. That's why the trilemma is usually solved by stacking layers, not by one perfect algorithm.
Chains that share a consensus family speak the same protocol dialect — a head start for interoperability. Different mechanisms mean a bridge must translate between two security models. Hover any cell.
A caveat: sharing a mechanism doesn't make two chains automatically interoperable — Ethereum and Cardano are both Proof of Stake yet still need bridges. But it removes one translation layer, while cross-consensus pairs (Bitcoin ↔ Solana) must always bridge between entirely different security assumptions.