InterPlanetary File System (IPFS)

Intermediate

Key Takeaways

IPFS is an open-source, peer-to-peer network protocol for storing and sharing files without relying on central servers.
Rather than locating files by server address (as HTTP does), IPFS locates them by their content, using a unique cryptographic fingerprint called a Content Identifier (CID).
IPFS is widely used as off-chain metadata storage for NFTs, decentralized applications, and distributed archiving projects.
Files remain accessible only as long as at least one node is "pinning" them; pinning services and protocols like Filecoin help address long-term availability.

IPFS (InterPlanetary File System) is an open-source protocol designed to create a distributed, permanent web where files can be stored and retrieved without depending on any single server. Created by Protocol Labs and first released in 2015, IPFS takes a fundamentally different approach to data distribution compared to the HTTP protocol that underpins most of the internet today.

HTTP vs. IPFS: Location vs. Content

The traditional web runs on HTTP, a protocol that retrieves content based on where it is stored. When you visit a website, your browser sends a request to a specific server at a specific address, such as an AWS data center, and that server returns the page. If the server goes offline, the content becomes unavailable.

IPFS inverts this model. Instead of asking "where is this file?", it asks "what is this file?" Each file on IPFS is identified by a Content Identifier (CID), a unique string derived from a cryptographic hash of the file's contents. Any node on the network that holds a matching file can respond to the request, regardless of its physical location.

This means that if you want to retrieve a document from IPFS, you’ll broadcast a request to the distributed network, and the closest nodes that hold the matching CID will respond. No single central server is required.

How IPFS Works

When you add a file to IPFS, the protocol breaks it into smaller chunks and builds a Merkle tree-style directed acyclic graph (DAG) of those chunks. Each chunk is hashed, and the hashes are combined to produce the root CID that uniquely identifies the entire file. This structure makes IPFS tamper-evident: changing even a single byte produces a completely different CID, so recipients can instantly verify data integrity.

File discovery and routing are handled by a distributed hash table (DHT), a lookup system that maps CIDs to the peer addresses currently hosting them. When a node requests a CID, the DHT finds which peers are pinning it and the file transfers directly between them, similar to how BitTorrent works but without any central tracker.

This content-addressed approach also integrates naturally with blockchain systems: a smart contract or on-chain record can store an immutable CID that points to off-chain data, providing verifiable links to large files without bloating the blockchain.

IPFS and NFTs

One of the most prominent real-world uses of IPFS is storing metadata and media files for NFTs. Because NFT smart contracts on blockchains like Ethereum reference an external URI pointing to the token's image and attributes, storing that data on a centralized server creates a risk: if the server goes offline, the NFT's metadata disappears. By routing to an IPFS CID instead, the metadata becomes content-addressed and verifiable, so any node pinning the CID can serve it.

This pattern is now standard across most NFT platforms. The on-chain contract stores the CID; services like Pinata or Filebase pin the underlying files to keep them available. The result is a tamper-proof link between the token and its associated media.

Availability and Pinning

A key characteristic of IPFS is that files do not persist automatically. A file is accessible only as long as at least one active node is "pinning" it. If all nodes that hold a file go offline, the file becomes unreachable, even though the CID remains valid.

To address this, a range of pinning services allows users and developers to pay for or incentivize continued file hosting. Filecoin, a blockchain-based storage network built by the same team behind IPFS, provides an economic layer for this: storage providers earn tokens in exchange for proving they continue to hold pinned data. This combination of IPFS (content addressing and retrieval) with Filecoin (storage incentives) is designed to make decentralized data persistence more reliable.

Now, tools like Kubo and the browser-native Helia client, along with service-worker gateways that let users access IPFS content directly from a standard browser without installing any software, have made the protocol significantly more accessible for developers and end users.

Advantages and Limitations

Potential advantages

Distributed availability: content can be served by any node holding it, reducing single points of failure compared to centralized hosting.
Data integrity: CIDs make it easy to verify that a file has not been altered.
Bandwidth efficiency: when multiple people request the same file, nodes can retrieve it from the nearest peer rather than a distant server, potentially reducing load and latency.
Blockchain compatibility: CIDs provide a compact, immutable reference to large off-chain files.

Limitations

Availability depends on pinning: if no node is actively storing a file, it becomes inaccessible.
Adoption is still growing: the network of peers is smaller than the centralized web, which can affect retrieval speed for rarely-requested content.
Complexity: Compared to HTTP, IPFS adds complexity for developers unfamiliar with content addressing and DHT routing.

FAQ

What is IPFS?

IPFS (InterPlanetary File System) is an open-source, peer-to-peer protocol for distributing and retrieving files. Unlike HTTP, which retrieves content from a specific server address, IPFS retrieves content based on what it is, using a cryptographic fingerprint called a Content Identifier (CID). Any node on the network that holds the matching content can serve it.

How does IPFS work?

When a file is added to IPFS, it is split into chunks, hashed, and assembled into a Merkle DAG. The root hash becomes the CID that uniquely identifies the file. A distributed hash table (DHT) maps CIDs to the peers currently hosting them. When you request a CID, the DHT finds the relevant peers and the file transfers directly between them, without going through a central server.

What is content addressing in IPFS?

Content addressing means that files are identified by a hash of their contents rather than by their location on a server. The resulting identifier, called a CID, is unique to that exact file. If the file changes in any way, the CID changes. This makes it possible to verify data integrity and retrieve the same content from any node that has it, wherever in the world that node may be.

What is the difference between IPFS and HTTP?

HTTP is a location-addressed protocol: it retrieves content from a specific server at a specific URL, so if that server goes offline, the content is gone. IPFS is a content-addressed protocol: it retrieves content based on a cryptographic fingerprint (CID) that any participating node can serve. IPFS is designed to be more resilient to server failures and better suited for distributed or long-term data storage use cases.

Closing Thoughts

IPFS represents a significant shift in how files can be stored and retrieved on the internet, replacing location-based addressing with content-based addressing. Its integration with blockchain systems and its role in NFT metadata storage have made it a practically relevant protocol within the crypto ecosystem.