Drawbacks of IPFS: A Brief Guide to Understanding IPFS & Where it Lacks

Before we dive into IPFS and its drawbacks, let’s discuss how the internet works and the common system architecture that is widely adopted across the world. If we define the internet as a huge network of interconnected computers, then the architecture of systems defines how the computers stay connected to each other.

Since time immemorial, we have been following the client-server architecture — which forms the basis for HTTP. In a client-server model, the data is stored in a certain number of servers from which information is accessed according to location-based addressing.

As the internet evolved from web1 to web2, and web2 to web3, the need to transfer huge amounts of data, be it audio or video content became commonplace. But HTTP wasn’t originally designed to facilitate huge magnitudes of data transfer. This led to the adoption of better and more efficient system architecture such as peer-to-peer architecture. Some of the earliest examples of applications built on a peer-to-peer architecture include Napster and Bittorrent. While IPFS was built on a similar architecture, Filecoin is an improvement on IPFS.

Read on as we dive deeper into the underpinnings of IPFS and its drawbacks.

What is IPFS?

InterPlanetary File System, short for IPFS, is an open-source file-sharing system that has been created by Protocol Labs by leveraging existing innovations in the realm of peer-to-peer systems. What this means is that information and data aren’t stored in a set of storage servers but are distributed across a number of systems (read peers). In essence, IPFS can be called a decentralized storage network. One of the biggest differences between IPFS and HTTP is the fact that the former relies on content-based addressing while the latter relies on location-based addressing.

According to the IPFS documentation, here are the three fundamental principles you should know to understand IPFS:

• Unique identification via content addressing: Like we just mentioned above, IPFS fetches data based on what the content is, but not where it is located.
• Content linking via directed acyclic graphs (DAGs): Content linking is done through a data structure known as DAG; Merkle DAG to be specific.
• Content discovery via distributed hash tables: For discovering content that is already in the network, distributed hash tables (DHTs) are used. While a hash table is a database that contains keys to various values, in a DHT, the table is split across all the peers in the IPFS network.

To learn more about each of these principles that govern the IPFS network, check out the documentation here.

Key Components of IPFS

In the previous section, we provided a quick overview of what IPFS is and its underlying principles. Now, we dive a little deeper into the components of IPFS. You can find out how the above mentioned principles tie into the various components in IPFS below:

• Distributed Hash Tables
  Distributed hash tables are a key part of IPFS as they help make the file sharing network decentralized, scalable, and fault-tolerant. This is because distributed hash tables spread data (which is in key/value pairs) across a huge network. DHTs also facilitate reliable access and searching among various nodes.
• Block Exchanges
  File sharing systems such as IPFS need to be able to handle transfer of huge amounts of data among millions of nodes. As a result, a strong data exchange protocol is indispensable. By the time IPFS was in development, Bittorrent was already able to create a data exchange protocol that can successfully coordinate upload and download of data among several million peers. IPFS implemented a spinoff of this exchange protocol and called it “BitSwamp.”
• Merkle DAG
  Merkle DAGs are a combination of a Merkle Tree and a Directed Acyclic Graph implemented together. Directed Acyclic Graphs are a data structure used for modelling and structuring data. They indicate transactions through various vertices that connect the nodes in the data structure, similar to how blocks indicate various transactions in a blockchain.

Challenges with IPFS

Sure, IPFS is a decentralized file sharing system that facilitates seamless, secure, and low latency data distribution. Unlike HTTP, it can be used for sharing large files over the internet. But it does come with its own share of limitations.

Lack of Strong Economic Incentives

IPFS, on paper, seems like the best alternative to HTTP for secure, high throughput transactions. But to actualize as a reliable, everyday solution and serve as a database-as-a-service model, it needs to create strong economic incentives. Since the start, IPFS has been built as a community product that will benefit from everyone’s contributions. As a result, no economic incentives were put in place. While lack of economic incentives do not weaken the technology or what IPFS aims to do, they simply make it an impractical solution for long-term use; especially for storing private and enterprise data.

Unreliable with Private Data

Be it Bittorrent or IPFS– both are perhaps suitable in a voluntary-collaborative space where academic research or sharing a particular type of data (be it music, movies, or books) is the need. This is because IPFS is inherently free and the peers/nodes in the network are not volunteering to store a huge magnitude of private data on their systems for free for a long time. For peers in the IPFS network to facilitate huge storage capacity for a sustained period, an economic incentive model needs to be in place. This isn’t the case with IPFS. Furthermore, storing encrypted data on a peer to peer network such as IPFS means it is hard to de-duplicate or cache efficiently

Lack of on-chain proofs:

With IPFS, along with not having an incentive layer, it is close to impossible to verify the integrity of the data that it stores. Peers do not have to submit proofs that they are indeed storing the data or its uptime. Currently, this is being solved by paying centralized gateway providers to pin your data, but this leads to a central point of failure and defeats the purpose of using a decentralized storage provider.

Filecoin & IPFS

Filecoin is an innovation that sprang from the applications of IPFS. What it does differently is that it introduces an economic model and incentives to IPFS. In simpler terms, Filecoin can be considered an electronic currency or crypto currency, much like Bitcoin. Users can simply provide a portion of their unused storage on their harddrives in exchange for currency.

While Filecoin’s efforts try towards making IPFS a reliable file sharing and storage platform, it is still plagued by some issues. We’ll expand more on Filecoin, what about it works and what doesn’t, in a later blog post. Follow us on our socials and subscribe to our email list on Medium for all our latest blog updates.

Alternatives to IPFS

An ideal alternative to IPFS would be a file sharing and storage system that is decentralized, highly secure, and has a strong economic incentive model. Some of the known alternatives to IPFS include Filecoin, Storj, and Arweave. As we have already discussed Filecoin above, we will move on to Storj and Arwaeave.

Storj

Storj is another decentralized cloud storage network that distributes each file you upload on the network into 80 pieces across its global servers. What this means is that you get to enjoy redundancy against server failures, fast download speeds, and solid security. That being said, Storj, when compared to traditional cloud providers such as Amazon or Azure, does not provide a simple user interface. You would need programmers to configure Storj as it is largely dependent on a command line interface. This becomes a major drawback for Storj as it doesn’t have an intuitive interface that can be used by just about anyone.

Arweave

Arweave emerged as a protocol that offers permanent storage. It is worth noting that other cloud providers, be it centralized or decentralized, do not offer such permanent storage options. Arweave also charges a fortune for the same, as it currently charges $5 for every GB stored on its network. In comparison, Amazon only charges $0.2 per GB. Although Arweave’s proposition of permanent storage is a novel one, the protocol as a whole, especially due to its steep pricing, is not a practical option for developers and business owners looking for a viable storage solution.

Arcana Network

Arcana Network is a decentralized storage network that incentivizes users through the $XAR token, and is optimized for storage of DApps and private data. While services such as Filecoin, Arweave, and IPFS are not a practical solution and strong competition to the traditional cloud storage market, Aracana is.

Arcana offers Social Auth, Access Management, and KMS in an easy to use SDK that takes minutes to integrate for any developer. Check out the testnet on testnet.arcana.network

About Arcana Network

Arcana Network is a decentralized storage layer for Ethereum, offering storage for DApps built on EVM compatible chains, such as Ethereum, Binance chain, and Polygon (Matic). But Arcana doesn’t stop at storage. To fully realise the privacy and data ownership benefits of decentralized storage, you need a suite of services, which are currently not decentralised. Arcana fixes this with its Privacy Stack. Arcana’s Privacy stack offers Decentralized Storage that is end-to-end encrypted, along with Non-custodial Key Management Services (KMS) and Decentralized Identity and Access Management.

Official Links: Website | Telegram | TG Announcement | Twitter | Medium