A node is any computer that participates in a blockchain network by storing a copy of the ledger and communicating with other participants. Validators are a specific type of node that actively propose or approve new blocks, making them the engines that keep a blockchain moving forward. Understanding what these machines do — and why it matters how many of them exist — is fundamental to understanding why blockchains behave differently from a traditional database controlled by a single company.
What a Node Actually Does
When you send a cryptocurrency transaction, it does not go to a central server. It is broadcast across a peer-to-peer network of nodes, each of which receives the message, checks whether it looks valid, and passes it on to its neighbors. No single machine is in charge; the network reaches agreement through rules that every participant follows independently.
A full node stores the entire history of the blockchain from the genesis block to the present. It independently verifies every transaction and every block against the protocol’s rules — it does not trust anyone else’s word. If a miner or validator tries to slip in an invalid transaction (say, spending coins that do not exist), full nodes reject it. This is the fundamental check on power in a decentralized network: you cannot fool thousands of independently operating computers all running the same rulebook.
A light node (sometimes called a simplified payment verification, or SPV, node) stores only block headers rather than full transaction data. It relies on full nodes for detailed verification. Most mobile wallets operate this way — they sacrifice some trustlessness in exchange for not requiring gigabytes of storage on your phone.
Why Node Count Matters
If only ten computers ran a blockchain, a government, corporation, or attacker would only need to coerce or compromise those ten machines to shut the network down or rewrite its history. With tens of thousands of nodes spread across many countries and operators, no single entity controls enough of the infrastructure to do that unilaterally. This geographic and organizational distribution is what people mean when they say a blockchain is censorship-resistant.
The number of nodes in a network is one of the most honest signals of its decentralization. A chain with a hundred validators controlled by a handful of related companies is functionally more centralized than it appears on paper.
Validators: Who Writes the Next Block
While every full node verifies blocks, only validators (sometimes called block producers, miners, or witnesses depending on the network) actually assemble new transactions into blocks and submit them for the network to accept.
How a node earns the right to produce a block depends on the consensus mechanism the network uses.
Proof of Work
In proof of work networks like Bitcoin, any node can become a miner by performing computational work — repeatedly hashing data until it finds a result that meets a difficulty target. The first to find a valid hash earns the right to add the next block and collect a reward. Becoming competitive requires significant hardware investment, which limits how many active miners exist, but the open entry means anyone with the resources can join.
Proof of Stake
In proof of stake networks like Ethereum (post-Merge), validators are chosen to propose blocks roughly in proportion to the amount of cryptocurrency they have locked up — “staked” — as collateral. If a validator tries to cheat, the protocol can destroy part of their stake through a process called slashing. This economic penalty is the primary enforcement mechanism: dishonesty becomes directly expensive.
Different proof-of-stake networks vary in their minimum stake requirements and committee sizes. Ethereum requires 32 ETH to run a solo validator, a threshold high enough that many participants delegate their stake to pooled services rather than running a validator themselves. This trade-off between accessibility and decentralization is an ongoing design tension across the industry.
The Validator Set and Decentralization
| Network Type | Who Can Validate | Typical Validator Count | Trust Model |
|---|---|---|---|
| Permissionless PoW | Anyone with hardware | Thousands of miners | Economic competition |
| Permissionless PoS | Anyone meeting stake threshold | Hundreds to thousands | Economic stake + slashing |
| Delegated PoS | Token holders vote for delegates | 21–100 elected nodes | Representative delegation |
| Permissioned / consortium | Approved organizations only | Tens | Institutional agreement |
The table above illustrates a spectrum. Delegated proof-of-stake systems like some enterprise chains intentionally trade breadth of participation for throughput and simplicity. Whether that trade-off is acceptable depends on what the network is trying to achieve.
A validator set that looks large can still be concentrated if many validators are run by the same legal entity or share infrastructure at the same cloud provider. A regional outage at a major cloud platform has, on multiple occasions, noticeably degraded networks whose validators were clustered in the same data centers.
Running Your Own Node
Anyone can download the software for most public blockchains and run a full node. Doing so does not typically earn rewards — you are providing a public service to the network, not producing blocks — but it gives you something valuable: you verify everything yourself. When you use an exchange or a third-party wallet, you are trusting their node. When you run your own, you are trusting the math.
The practical requirements vary widely. Bitcoin’s full node software runs comfortably on modest hardware with a few hundred gigabytes of storage. Running an Ethereum full node requires more disk space and memory, and archival nodes (which keep all historical state) can require multiple terabytes. Some networks are designed to keep node requirements deliberately low to maximize participation; others accept higher requirements as the price of greater throughput.
If running a full node is out of reach, understanding what nodes do still matters for how you evaluate a project. Questions worth asking: How many independent validators does the network have? Are they geographically distributed? Is the validator set open to new entrants, or controlled by a fixed group? These questions cut through marketing to the actual security model.
Key Takeaways
- A full node stores the complete blockchain and independently verifies every transaction — it trusts no one else.
- Validators are the nodes that produce new blocks; how they are selected depends on the consensus mechanism.
- In proof of work, computational effort grants the right to mine; in proof of stake, locked collateral (with slashing risk) grants the right to validate.
- The number and diversity of nodes is a direct measure of a network’s decentralization and censorship resistance.
- Running your own node is the strongest form of self-sovereignty — you verify the chain yourself rather than relying on a third party.
- Validator concentration (by entity, geography, or infrastructure provider) is a real risk even in nominally decentralized networks.
Next up: Cryptographic Hashing