How Does Walrus Differ From Existing Storage Solutions?

After the successful launch of Sui Network and the Deepbook protocol, Mysten Labs is now venturing into a new, highly promising field with Walrus. The success of Sui and Deepbook has generated significant anticipation for Walrus. However, there are still concerns surrounding this new protocol. These concerns stem from various factors such as: the decentralized storage market being saturated with many solutions that are not truly optimized, and worries about resource allocation — specifically whether Mysten Labs, which is focusing heavily on developing and expanding Sui Network, is spreading its resources too thin while pursuing new initiatives. The article by Four Pillars will explain the structure of Walrus Protocol and explore how it differs from existing decentralized storage solutions. Then, it will analyze the relationship between Walrus and Sui Network, focusing on how Walrus integrates with Sui’s architecture and contributes to the expanding Sui ecosystem. How is Walrus different from existing storage solutions? To answer why Walrus was created, we first need to examine its differences compared to current decentralized storage solutions. Walrus stands out from current storage models (especially Filecoin and Arweave) in three main aspects: Cost Optimization Firstly, the storage costs among Walrus, Arweave, and Filecoin vary significantly.

• Arweave uses an incentive system encouraging nodes to replicate and store as much data as possible.
• Filecoin allows users to decide how many nodes will store their data (users can choose to store data with just one miner or distribute 100 copies across 100 different miners, of course, the more copies required, the higher the cost).
• Walrus employs Red-Stuff encryption, which has demonstrated lower costs and up to 100 times higher efficiency compared to Arweave and Filecoin (compared to Arweave, which requires storing data across the entire network leading to higher replication costs up to 500 times, Walrus maintains efficiency with only 4-5 copies). At the same time, the risk of data loss is significantly reduced. Simply put, Walrus addresses the limitations of both Arweave and Filecoin. Arweave has a low risk of data loss but high replication costs. Conversely, Filecoin offers storage services at more reasonable costs based on user needs, but the risk of data loss increases with cheaper packages. Walrus combines the advantages of both, maintaining low replication costs while minimizing data loss risk. Additionally, as the number of nodes on Arweave increases, costs also rise (mildly) because the incentive system encourages all or only designated nodes to store as much complete data as possible. In contrast, Walrus only needs to transmit data once across the entire network. Each node stores only a part of the data. This significantly reduces the load on individual nodes as the network expands. This structural difference allows Walrus to optimize storage costs more effectively than Arweave and Filecoin. Programmability While Walrus’s cost efficiency compared to Arweave and Filecoin is a major advantage, the “programmability” feature is what truly sets it apart. If traditional storage solutions are merely data repositories, Walrus enables decentralized storage systems to be programmable via Sui Network, opening up functionalities far beyond basic data storage. What if smart contracts could directly reference or trigger data within the decentralized storage system? For example: when minting an NFT, image files could be stored on Walrus, and their blobs could be created on Sui Network, creating a seamless link with the NFT. This addresses common criticisms of traditional NFTs being “incomplete” (with on-chain tokens but off-chain stored metadata). With Walrus, NFTs can truly become complete Web3 assets. Another example of Walrus’s potential is that its data blobs can be stored as Sui objects and controlled via Sui’s Move smart contracts, allowing data to be transferred to other users or ownership to be automatically changed. This is why we assert that data on Walrus can be programmable. Conversely, Arweave and Filecoin are limited and almost incapable of dynamic on-chain integration. Although Filecoin has added some smart contract functionalities via FVM (Filecoin Virtual Machine), editing and controlling data remains limited. This makes Walrus significantly superior in programmability. Data Access and Deletion Traditional storage protocols share a common trait: once data is uploaded, anyone can access it, and it cannot be deleted. While this feature may be useful for individual users, it poses challenges for organizations and enterprises that require data security or need to modify/delete data. Walrus solves this by allowing users to delete or modify data when needed (which is very different from Arweave, where data is “frozen” permanently, and Filecoin, where data deletion occurs not by user choice but when contracts expire or storage nodes encounter issues and cease to exist on the network). Some may worry that this contradicts the immutability principle of blockchain, but it’s important to note that Walrus only handles blob data, while transaction data remains immutable, ensuring blockchain integrity. With enhanced practicality compared to traditional storage systems, Walrus has great potential for adoption by traditional enterprises and Web2, leading to high user expectations for its future flexibility. How do Walrus and Sui Network work together? After understanding the differences between Walrus and traditional storage protocols, let’s explore the relationship between Walrus and Sui. When Mysten Labs announced plans to launch Walrus Protocol, many expressed concerns and asked: “Shouldn’t they focus entirely on Sui instead of creating another protocol?” However, a brief understanding of how Walrus operates clearly shows that Walrus does not distract from Sui’s focus; in fact, Walrus should be viewed as a storage layer built to complement applications on Sui. Walrus not only adds storage capabilities to Sui but also positively impacts SUI, the governance token of Sui Network, making it impossible to see Sui and Walrus as two separate systems. Supportive Relationship between Sui and Walrus In fact, Mysten Labs recognized early on the importance of managing data storage during Sui’s initial design phase. Data on the blockchain inevitably grows with increasing transactions, which can lead to higher transaction fees and negatively impact the user experience on Sui in the future. Therefore, from the early design stage, Mysten Labs introduced a unique concept called Storage Fund (Storage Reserve) to address Sui’s storage challenges. The Storage Fund on Sui works as follows: When users send transactions to Sui validators, transaction fees are split into two parts: 1)gas fees for computation and 2)storage costs. Sui collects storage costs upfront from users to store data permanently and pools this amount into the Storage Fund. The Storage Fund then continuously distributes the accumulated funds to validators while data remains stored on-chain. Additionally, users can be refunded for storage costs if they delete their data. This unique data storage system on Sui creates two key effects: Users can be refunded for storage costs when deleting on-chain data, encouraging everyone to reduce ledger size through economic incentives. The reward structure, which collects storage fees upfront and allocates them as future validator rewards, can address sustainability issues related to storage. Although Sui has an effective mechanism to handle storage, directly storing large blob data files (such as multimedia files) on-chain remains challenging. This is where Walrus comes in — it allows large data files to be stored off-chain while creating metadata on Sui to control them, enabling programmable data without direct on-chain storage. Furthermore, through Sui, Walrus leverages its most distinctive feature compared to other storage protocols: programmability and data control. Ultimately, Sui and Walrus form a supportive relationship, creating unique advantages and complementing each other’s limitations. Walrus Turns SUI into a Deflationary Asset As mentioned in the Storage Fund example, Sui Network requires users to pay a certain amount of tokens to store any object on the blockchain. Walrus also follows this principle. When creating blobs on Walrus, an amount of SUI tokens equivalent to the object’s size (here, the size of the object representing the blob, not the actual blob size) will be locked in the Storage Fund. Although users can be partially refunded when deleting data, part of the fee will be burned, permanently removing tokens from circulation. This means that the more data stored via Walrus, the more tokens are permanently locked in the Storage Fund, creating a positive feedback loop: as Walrus usage increases, the token supply decreases. In summary, Walrus’s emergence offers significant benefits for Sui, both in terms of network and asset value. Walrus is expected to open up more diverse development directions for the Sui ecosystem. Walrus as the “Most Important Piece” in the Sui Ecosystem Mysten Labs: Building a comprehensive Web3, not just blockchain Initially, the author thought Mysten Labs was solely focused on developing Sui. But after witnessing the emergence of Deepbook and Sui Naming Service, the author began questioning Mysten Labs’ ambitions. Until Walrus was launched, the author concluded that they are aiming to build a complete decentralized infrastructure for Web3. What makes Mysten Labs different is their long-term vision and methodical approach. They do not chase short-term token issuance trends but focus on creating innovative solutions across all aspects: execution, storage, consensus, and communication. They also understand Web2 user psychology and strive to deliver the most user-friendly experience. Specifically, Sui Network handles execution and consensus (continuously upgraded via Mysticeti and Pilotfish & Remora), Walrus is responsible for storage, and SCION (the next-generation internet architecture), capable of protecting network packets, DDoS resistance, and routing attack immunity — although not developed by Mysten Labs, it will be applied across the entire Sui Network(, ensuring secure communication, with tools like zkLogin, Stashed, SEAL, and KELP providing familiar interfaces for Web2 users. If these pieces are well integrated, the author believes Mysten Labs will redefine the current Web3 model. They are not just a blockchain company; they are building infrastructure for a completely new “web.” Sui is at the core of this vision, and Walrus is the most crucial “piece.” Walrus Is Not Limited to the Sui Ecosystem However, Walrus is not confined to the Sui ecosystem. Like other storage protocols, Walrus can be used by any third party outside Sui applications. It has the potential to become a useful alternative to current storage protocols or even Data Availability layers such as Celestia, EigenDA, and Avail. This usability of Walrus will expand the demand for SUI tokens beyond Sui Network. When Walrus is used, objects are created on Sui Network, leading to a reduction in SUI supply. In other words, Walrus has the potential to make SUI tokens more attractive by creating external demand. Therefore, Walrus is expected to serve as a bridge to expand Sui in multiple directions. Will Walrus surpass Filecoin? While comparing the value of specific protocols is delicate, the author is optimistic about Walrus’s future for the following reasons:
• Walrus has a superior operational mechanism compared to current storage protocols.
• Walrus can perform tasks that existing storage protocols cannot, such as becoming a Data Availability layer or enabling programmable stored data.
• Walrus benefits from the strong network and user community of Sui. If Walrus not only becomes a storage layer for Sui but also a storage protocol representing Web3, aligning with Mysten Labs’ vision, it could potentially lead the storage field.