Recently, blockchain storage has become active again. Filecoin is working on the F3 consensus upgrade, Arweave is expanding its ecosystem roadmap, while IPFS hasn't gained much attention. Suddenly, Walrus emerged out of nowhere, and less than a year after its mainnet launch, it has gained support from several top exchanges, with community enthusiasm skyrocketing. I took some time to review its technical approach and found that it indeed has unique features.
Let's start with the underlying coding. Walrus uses RedStuff erasure coding, which is fundamentally different from traditional full-replica storage. It slices files into multiple slivers, encodes them with Reed-Solomon algorithms, and disperses them into shards across various storage nodes. The key point is—only 1/3 of the symbols are needed to fully recover the entire file. What does this mean? Suppose there are 300 storage nodes; even if 200 nodes fail or act maliciously, your data can still be fully retrieved.
Looking at Filecoin's Proof of Replication (PoRep) scheme, it also guarantees data integrity, but fundamentally it still requires storage providers to keep a complete copy. It uses cryptographic verification to ensure you're actually storing the data, preventing forgery, but the redundancy overhead is significant—storing 1TB of data might require several TBs of physical storage on the network. Walrus only needs five times redundancy to achieve the same security level, which clearly offers better efficiency.
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.
18 Likes
Reward
18
7
Repost
Share
Comment
0/400
LiquidatedAgain
· 01-21 17:20
It's hard to buy early knowledge, Walrus's erasure coding is indeed powerful—restoring data with just 1/3 of the data. The redundancy overhead of Filecoin is outrageously high.
Once again, I see storage projects being liquidated. I now have to consider risk control points when looking at storage projects. With how fast Walrus is listed on exchanges, the lending rates probably won't be off the charts.
Filecoin spends a lot of money maintaining those copies, but five times redundancy with Walrus is more cost-effective. But don't tell me to bottom fish—I’ve lost too much money on that.
Less than a year after the mainnet launch, top-tier exchanges are supporting it. The hype is really ramping up—be cautious of leverage temptations.
RedStuff algorithm is awesome, but I just want to know if the cost of malicious nodes is high enough to withstand real-world application tests.
Filecoin has been around for so many years, still needing to keep complete copies. Walrus, on the other hand, surpasses it—this gap is truly outrageous.
The storage sector has suddenly heated up again. IPFS has no buzz but has become a spectator. Who can make it to the end in this round? It’s really hard to say.
View OriginalReply0
GateUser-40edb63b
· 01-18 19:05
Walrus this move is indeed impressive. Erasure coding saves much more storage space than Filecoin. With 5x redundancy, it can do the work of Filecoin. No wonder exchanges are competing to list it.
View OriginalReply0
MEVHunterBearish
· 01-18 17:48
Walrus's technology is indeed excellent. One-third recovery is much smarter than the Filecoin system, with less redundancy and higher efficiency. This is what the future of storage should look like.
View OriginalReply0
BridgeJumper
· 01-18 17:41
Walrus's technology is truly excellent; 5x redundancy outperforms that old-fashioned Filecoin solution.
View OriginalReply0
AirdropGrandpa
· 01-18 17:34
Walrus's technical approach really has some tricks up its sleeve; this erasure coding system is truly more advanced than full replica storage.
View OriginalReply0
SmartContractRebel
· 01-18 17:31
This Walrus is really amazing. The erasure coding approach is much smarter than Filecoin's replication scheme. Five times redundancy beats their dozens of times, no exaggeration.
---
After all these years, Filecoin is still digging the ground, but Walrus came out and immediately secured top-tier exchanges. That's interesting.
---
Basically, the Reed-Solomon algorithm crushes full replication. Even with 200 nodes crashing, it's fine. The security is indeed unmatched.
---
Wait, can it recover with just 1/3 of the symbols? If the data volume is large, the storage cost would be incredibly low. No wonder the community is so enthusiastic.
---
IPFS really needs to reflect. With Walrus's interference, it has lost all momentum, haha.
---
The efficiency of redundancy is so poor. Will Filecoin users start considering migration? That’s the real threat.
View OriginalReply0
CoffeeNFTrader
· 01-18 17:24
Walrus's technical approach is indeed impressive; 5x redundancy beats the Filecoin system.
Recently, blockchain storage has become active again. Filecoin is working on the F3 consensus upgrade, Arweave is expanding its ecosystem roadmap, while IPFS hasn't gained much attention. Suddenly, Walrus emerged out of nowhere, and less than a year after its mainnet launch, it has gained support from several top exchanges, with community enthusiasm skyrocketing. I took some time to review its technical approach and found that it indeed has unique features.
Let's start with the underlying coding. Walrus uses RedStuff erasure coding, which is fundamentally different from traditional full-replica storage. It slices files into multiple slivers, encodes them with Reed-Solomon algorithms, and disperses them into shards across various storage nodes. The key point is—only 1/3 of the symbols are needed to fully recover the entire file. What does this mean? Suppose there are 300 storage nodes; even if 200 nodes fail or act maliciously, your data can still be fully retrieved.
Looking at Filecoin's Proof of Replication (PoRep) scheme, it also guarantees data integrity, but fundamentally it still requires storage providers to keep a complete copy. It uses cryptographic verification to ensure you're actually storing the data, preventing forgery, but the redundancy overhead is significant—storing 1TB of data might require several TBs of physical storage on the network. Walrus only needs five times redundancy to achieve the same security level, which clearly offers better efficiency.