How to Run a Polygon Node: Requirements 

Polygon is a blockchain network, designed as the Ethereum scaling solution, which addresses the problems of its high gas fees and low transaction bandwidth. It works on the Ethereum Virtual Machine (EVM) and it is compatible with Ethereum and its other layer-2 (L2) solutions using bridges.

Let’s dive deeper into its functionalities and start with the definition of its nodes.

This guide was first released in July 2024. It’s been revised to reflect the ecosystem changes and the most up-to-date Polygon full node requirements.

What is a Polygon node?

A blockchain node is a computer server that runs blockchain software, stores its transactions, and proceeds with requests. Hence, the Polygon node is the server that runs Polygon’s source code. To utilize it, users need an RPC endpoint, which enables them to interact with the blockchain using the network methods.

Source: GetBlock

GetBlock offers a variety of Polygon (MATIC) RPC nodes for different purposes, including shared and dedicated nodes.

What is a Polygon (MATIC)?

Polygon is a fully EVM-compatible Ethereum scaling solution, while MATIC used to be its native token. In September 2024, the network transitioned to POL as its main token. However, despite the change, MATIC may still appear on older materials and interfaces.

Polygon tokens are stored on addresses which look similar to Ethereum addresses, and they can be swapped with Ethereum and other L2 using bridges and paying gas fees. It has its own developed infrastructure, with DeFis, dApps, tokens, and NFTs.

Types of Polygon nodes

The Polygon node structure is quite unique compared to other networks, as it has different node types: Heimdall, Bor, sentry, and validator nodes. They all are used to maintain the network and its connection with Ethereum. Let’s look closer at Polygon node peculiarities and then requirements.

What is a Heimdall node?

Heimdall is the Polygon validator layer, so its users can validate the state of the network, earning rewards. They use this Polygon node and stake POL tokens on Heimdall’s staking module, which then sends transactions to the Ethereum network, validating it and ensuring its functionality.

That’s how Polygon realizes its mission as the Ethereum scaling solution, ensuring its functionality while expanding its capabilities.

What is a Bor node?

Bor is another specific Polygon layer which is responsible for block production. According to the Polygon Foundation, validators who run a Bor Polygon node are eligible to participate in this process, with the probability of being chosen proportional to the amount of POL they stake.

Therefore, both Heimdall and Bor nodes are necessary to validate the network. Let’s move on and see how they are used.

What is a Polygon full node?

A full node contains all transactions that have ever been made since the blockchain network’s launch. It’s usually very large in terms of file size and computational resources needed to proceed with it, and as the blockchain is used, the full node becomes larger and larger.

To address resource constraints, the ecosystem offers several variations of full nodes:

• Pruned node: To avoid overcomplication, blockchain users can deploy a pruned node, which is essentially a full node with the most irrelevant information being removed. Usually, it’s enough for developers and blockchain researchers.
• Polygon archive node: Unlike a pruned node, an archive node contains all data that was recorded in the blockchain, from the very beginning of the network, while updating after each transaction. Therefore, it can be used to validate the Polygon network state at any given time moment.
• Sentry node: A sentry node is a full node with both Heimdall and Bor nodes software installed, so it can participate in the Polygon chain validation and new block creation. It’s the required component of the validator node that will be discussed below: a node that contributes to the whole network’s operation.

A short comparison of Polygon nodes is provided below.

Node type	Node structure	Node purpose
Simple (light) node	Contains the Polygon blockchain software and the latest transactions in the network, while earlier ones are validated using full nodes.	Routine blockchain operations, dApp development
Heimdall node	Contains specific software for maintaining the Polygon’s Proof-of-Stake mechanism and sync it with Ethereum.	Polygon staking and validation management
Bor node	Contains specific software for maintaining the Polygon’s consensus mechanism and produce new blocks.	Polygon consensus and new block production
Pruned node	Contains the most relevant and essential historical data about the Polygon transactions.	A semi-full node with the irrelevant archive information removed
Full (archive) node	Contains all historical information about the Polygon blockchain from the first transaction.	A full node that contains all blockchain data from the very beginning
Sentry node	Contains everything present in the full node, plus Polygon Bor and Heimdall nodes’ instruments, thus being capable of both Polygon validation and voting for the new block production.	A full node with both Bor and Heimdall nodes deployed, used for Polygon chain validation

Polygon validator node

As one can see, Polygon have different layers and nodes to maintain its performance, but what is a validator node? It is similar to a sentry node, with both Bor and Heimdall running on it, but it operates as a connection with Ethereum. So, it uses Ethereum RPC, rather than Polygon, and receives requests from a sentry node to maintain the network state.

Source: Polygon Foundation

Being a validator requires extensive preparation, which is described in detail in official guidelines. Validators stake POL and send the current Polygon’s network state to Ethereum, ensuring that both networks operate properly. They can set commission rates and obtain POL tokens as a reward for their activities.

Importance of running a Polygon node

The Polygon light node is enough for most of the development and exploring purposes. So, it’s no reason to launch a full node unless a user wants to dive deep into the blockchain history for research purposes or become a validator. Developers can use a light or pruned node instead, based on the amount of information they need for their applications.

Still, as one will see in the next questions, it requires robust hardware and high software literacy. While you can run a Polygon node by yourself, it may be a much better option to use the node provider service.

Polygon node requirements

If you need access to Polygon and decide to run node on your own infrastructure, you must account for compute, networking, and storage from day one. As your RPC provider, we take care of all of these operational details for you. Below is a concise summary of the official minimum specs that every full node must meet.

Hardware and software requirements for Polygon Full Node

Running the full node requires extensive computational capacities, as it means storing and proceeding with the large amount of data. The official Polygon Foundation requirements are at least:

• 8 CPUs/cores;
• 32 GB of RAM;
• from 4 to 6 terabytes of SSD memory.

Note, that since our original July 2024 post, the minimum specs for a Polygon full node have roughly doubled. As the network usage grows, every block adds more transactions, logs, receipts, and contract data. So for the chain in its current state, nodes need more storage, CPU power, and memory bandwidth.

Network and bandwidth requirements

Polygon mainnet full nodes require a minimum 1 Gbit/s network connection. For Amoy testnet, the recommendation is the same 1 Gbit/s interface, though overall storage/RAM needs are lower.

Keeping at least a gigabit link ensures your node stays fully in-sync, responsive to high-throughput dApps hitting your node endpoints, and ready for production workloads.

Polygon validator node requirements

The official Polygon requirements for becoming a validator can be summarized as follows.

• Meeting all hardware requirements for running a full node.
• Having two machines for running full nodes: for sentry and validator node.
• Opening and configuring machines’ ports.
• Having required RPC endpoints: Ethereum for validator, and Polygon for sentry.
• Having Polygon snapshots downloaded.
• Installing the RabbitMQ software for node management.
• Installing Bor and Heimdall services on both machines.
• Configuring seed nodes (seeds) for initial peer discovery.
• Following the validator’s requirements and best practices provided by the Polygon Foundation.

There aren’t specific requirements for the POL and ETH amount necessary to become a validator: they just need at least 1 POL and enough ETH to pay gas fees.

How to set up and run the Polygon node

Whether you want to implement blockchain in your app or become a validator, you need a node. You need the RPC MATIC endpoint and the coding environment to insert blockchain requests using the network methods. While you can use the node provider to make it easier for you, it’s still important to use blockchain methods properly when running the node.

Step-by-step setup guide

Thus, for the most cases, such as blockchain app development, you don’t need a full node that you deploy by yourself. So, let’s see an action algorithm that must be deployed to run Polygon node using the GetBlock access token.

1. Get a GetBlock Polygon mainnet endpoint in the GetBlock user dashboard to have the access token. You can also use the MATIC testnet RPC and obtain test tokens via our faucet but remember that they have no real-world price and value.
2. Make an RPC call using your access token and the Polygon methods. Use our Postman workspace guide to understand how to make calls.
3. Check Web3 libraries, such as Ethers.js and Web3.js, to implement automated RPC calls in your application where necessary, finishing the node implementation.

After the Polygon node setup, you can use it for blockchain testing, running your own dApp, more efficient Polygon tokens trading, or POL staking.

Syncing and configuring your Polygon node

If you point your dApp or service at GetBlock’s Polygon RPC endpoints, you don’t have to run, sync, or tune your own node at all. GetBlock maintains a fleet of fully-synced, highly available Polygon full nodes and archive nodes if you subscribe to that tier.

So your only job is to sign up, generate an endpoint URL, and configure your client. For instance, in your environment (e.g. .env), set:

POLYGON_RPC_URL="https://go.getblock.io/"

How to make RPC calls?

An RPC call is how you connect with the network to retrieve some information from it, write it into the blockchain, sign a smart contract, or perform another function provided by the network. That’s how you run nodes.

To make a Polygon RPC node call, you need to have the node endpoint, which you can obtain at GetBlock for free. Once connecting it to your wallet, such as MetaMask, you can use the eth_accounts method to retrieve the wallet addresses connected to the endpoint.

Would you like to know how to add GetBlock endpoints to MetaMask? Check our guidelines!

As you have your Polygon MATIC node endpoint, use the following script in your code:

curl --location --request POST 'https://go.getblock.io/' \
     --header 'Content-Type: application/json' \
     --data-raw '{
       "jsonrpc": "2.0",
       "method": "eth_accounts",
       "params": [],
       "id": "getblock.io"
     }'

The result will be:

{
    "id": "getblock.io",
    "jsonrpc": "2.0",
    "result": []
}

Use the GetBlock Polygon network API reference to learn more about methods used in the Polygon network for different purposes, and feel free to experiment with them!

Costs of running a Polygon node

GetBlock is a top blockchain provider, with 99.99% uptime for most of its nodes and transparent pricing, where you pay only for requests you use for connecting the blockchain. We also have a welcome reward: a 20% discount for all users who purchase our shared nodes on the annual subscription.

Let’s look closer at the offer.

Estimated costs for running a Polygon node

Shared nodes can be used by multiple users, so their request speed is limited. While we provide 50,000 Compute Units (CU) for free each day with Polygon public nodes, at a 5 RPS speed, the true GetBlock potential is revealed when using our paid solutions. Starting from $39/month, it provides from 50 to 600 million requests per month, with up to 200 RPS speed, while unused CUs are automatically moved to the next month if you prolong the subscription.

You’ll have a $100 bonus when purchasing our most in-demand plan, which scales for the larger and custom plans.

Flexible pricing

A Dedicated Polygon node is essentially your personal server for communicating with the blockchain. As no other users have access to it, it can operate at maximum speed, being able to proceed thousands of requests instantly. Therefore, such nodes are indispensable for large-scale dApps and services, used by thousands of customers.

Their price starts from $1,000/month for a node or fleet of servers configured according to your business size and needs.

DEDICATED NODE

Private server with custom settings

Requests

Unlimited

Rate Limit

Unlimited

Data

Archive

Starting at

$

600

/ month

Start building

For large enterprise solutions, which handle tens of thousands of clients each day, we have enterprise solutions at a negotiable price. With such large-scale Polygon crypto nodes, a user obtains the customized infrastructure with unlimited request number and speed, and we also provide guaranteed service level agreements (SLAs).

Polygon nodes of different types can be used for a vast number of goals, from exploring and creating dApps to validating the blockchain and contributing to its growth and development. GetBlock has solutions for different developers, validators, traders, and blockchain enthusiasts, which scale along with their projects and activities. Choose the plan that works best for your Polygon RPC node now, and get a 20% benefit from the first month!

Cloud vs. on-premises hosting

When you decide to run a blockchain node, whether it’s Polygon, Ethereum, Solana, or anything else, you have two broad choices.

Cloud hosting means you rent virtual machines (VMs) or containers from a public cloud provider (AWS, GCP, Azure, etc.) and install your node software there. As an advantage, the cloud provider handles underlying hardware maintenance. Costs vary by provider and region, and you need a good handle on cloud option cost policy and blockchain-node patterns to really optimize your bill.

On-premises hosting implies you own or lease physical servers in a data center and run your node software on that hardware. While it gives full control over server settings, you also have to fund that initial hardware purchase plus ongoing expenses like power, internet, and maintenance.

In short, cloud vs. on-prem is a spectrum of control vs. convenience. By contrast, an RPC provider like GetBlock sits on the convenience end, offering managed, high-performance endpoints at predictable costs without the headaches of infrastructure management. Meanwhile, dedicated plans let you have control of self-hosting but with all the ops work outsourced.