What has the Walrus project been up to lately? Evolving from a dedicated storage solution in the Sui ecosystem to a foundational infrastructure for all of Web3, there are several hardware metrics behind it that are worth clarifying.
Let's start with the technical aspects. After adopting a 2D erasure coding scheme, the replication factor has been reduced to 4-5 times—compared to traditional storage solutions that require 25 to 1000 times. The magnitude of this difference needs no further explanation. Data recovery costs can be as low as O(B/n), and full network synchronization only consumes O(B) bandwidth, translating to approximately $50 per year for 1TB. For comparison: Filecoin is 75% more expensive, Arweave is 98.6% more expensive. Performance metrics now rival centralized cloud storage, especially friendly for large-scale Blob files.
The design of programmable storage is particularly interesting. Storage objects are directly equivalent to on-chain programmable assets, managed dynamically via Move contracts. NFT metadata can be updated in real-time, assets in blockchain games can be reused across scenarios, and copyright revenue sharing can be automated—improving efficiency by a factor of 30. This is not just storage functionality; it’s more like laying the foundational data infrastructure for the Sui ecosystem.
Their cross-chain ambitions are also clear: not limited to Sui, they aim to serve as a data layer for other public chain Rollups, capable of handling AI training data, high-definition media, and DApp frontends. Essentially, they want to become the public "storage + data circulation" infrastructure for Web3.
Looking at the implementation data: the mainnet has expanded from 7 epochs to 103 storage operators, with 121 nodes, a total staked amount of 1.01 billion WAL, over 53.7 million transactions, and 1.64 million active addresses. Market performance is also responding: a market cap of 5.601 billion RMB, a 24-hour increase of 16.78%, a trading volume of 295 million, and a 7-day increase of 31.32%. Data is growing, the ecosystem is expanding, and this pace is accelerating.
This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.
16 Likes
Reward
16
6
Repost
Share
Comment
0/400
TopBuyerForever
· 7h ago
Whoa, copying factor has been pushed to 4-5 times? This number is indeed absolute. It's much cheaper than Filecoin, so why hasn't it exploded yet?
View OriginalReply0
TokenomicsDetective
· 13h ago
50 bucks for 1TB, really impressive, but what do Filecoin and Arweave think about this?
---
Can programmable storage be 30 times more efficient? Can cross-scenario reuse of chain game assets really be implemented, or is it just another PPT promise?
---
Currently 103 nodes, can they support the underlying infrastructure of Web3? Feels like it's still a bit early.
---
A 16.78% 24h increase, this number is a bit risky, need to see if it can stabilize later.
---
Walrus wants to take a slice of the DA layer cake. What does Celestia say? Will the competition be fierce?
---
Staking over 1 billion WAL, is the risk diversification enough? Is the operator concentration high?
---
30 times efficiency improvement, this number is quite impressive. Can it be replicated in real scenarios?
---
From Sui-specific to the entire Web3 underlying layer, the ambition is not small, but what about the difficulty of ecosystem collaboration?
View OriginalReply0
SignatureDenied
· 01-09 16:53
50 bucks to store 1TB for a year? Is this number real... Can it survive being so much cheaper than Filecoin?
View OriginalReply0
CryptoDouble-O-Seven
· 01-09 16:52
Wow, Walrus's cost is pushed down to $50 a year? That's way cheaper than Filecoin, pretty impressive.
View OriginalReply0
0xOverleveraged
· 01-09 16:49
Damn, Walrus's replication factor is pushed to 4-5x? 75% cheaper than Filecoin, I'm directly shocked. Is this real?
Sui's data infrastructure is about to take off. Programmable storage is indeed interesting.
$50 per year for 1TB, feels like copying the centralized cloud storage approach... but this idea is truly brilliant.
103 operators, 1.64 million active addresses. The growth rate is quite aggressive.
Wait, is the 16.78% 24-hour increase being pumped up with these news, or is there real fundamental support?
Cross-chain DA layer sounds ambitious, but is it just another hype?
View OriginalReply0
ValidatorViking
· 01-09 16:42
ngl the 4-5x replication factor vs those 25-1000x numbers hits different... actually battle-tested compression or just theoretical flexing?
What has the Walrus project been up to lately? Evolving from a dedicated storage solution in the Sui ecosystem to a foundational infrastructure for all of Web3, there are several hardware metrics behind it that are worth clarifying.
Let's start with the technical aspects. After adopting a 2D erasure coding scheme, the replication factor has been reduced to 4-5 times—compared to traditional storage solutions that require 25 to 1000 times. The magnitude of this difference needs no further explanation. Data recovery costs can be as low as O(B/n), and full network synchronization only consumes O(B) bandwidth, translating to approximately $50 per year for 1TB. For comparison: Filecoin is 75% more expensive, Arweave is 98.6% more expensive. Performance metrics now rival centralized cloud storage, especially friendly for large-scale Blob files.
The design of programmable storage is particularly interesting. Storage objects are directly equivalent to on-chain programmable assets, managed dynamically via Move contracts. NFT metadata can be updated in real-time, assets in blockchain games can be reused across scenarios, and copyright revenue sharing can be automated—improving efficiency by a factor of 30. This is not just storage functionality; it’s more like laying the foundational data infrastructure for the Sui ecosystem.
Their cross-chain ambitions are also clear: not limited to Sui, they aim to serve as a data layer for other public chain Rollups, capable of handling AI training data, high-definition media, and DApp frontends. Essentially, they want to become the public "storage + data circulation" infrastructure for Web3.
Looking at the implementation data: the mainnet has expanded from 7 epochs to 103 storage operators, with 121 nodes, a total staked amount of 1.01 billion WAL, over 53.7 million transactions, and 1.64 million active addresses. Market performance is also responding: a market cap of 5.601 billion RMB, a 24-hour increase of 16.78%, a trading volume of 295 million, and a 7-day increase of 31.32%. Data is growing, the ecosystem is expanding, and this pace is accelerating.