Olivia
Online
imundong
@kekeundo
A Lonely Tomato in the Online World..
Joined August 2025
340 Following
173 Followers
155 Posts
@yjkang686510 ㄱ ㄱ ㅑㅑㅑㅑㅑㅑ 새우!!!!!!! 축하해
@soyouIJ 마싯겟다아
@moooo_iii Hehehehehe 🤍✨🪽
gMic!
I went to a poker event today. . .
I'm so bad at poker
I need to study how to play :3
@SeismicSys @NoxxW3
@leee876212 큰거온다 영어로 어케해요
@moooo_iii gmum moi 🤍✨🪽
Hiii fren! :3 gMum!
Today I spent a bit more time looking into the products Optimum is building.
After reading so much about RLNC and data propagation, I honestly thought Optimum was just a networking project. But the more I dug into it, the more it felt like they're aiming for something much bigger.
Right now there are three main products: mump2p, DeRAM, and DeROM.
At first the names all blended together in my head T.T but once I started reading about them individually, the differences became pretty clear.
mump2p is the easiest one to understand because it's where RLNC is actually being used.
Its job is to help blocks, transactions, and other data move through the network faster. What caught my attention is that it works alongside existing libp2p infrastructure rather than trying to replace everything.
It feels more like an upgrade to the way data spreads across the network.
It's also the only product that's available today, so a lot of the RLNC concepts I've been reading about connect directly to mump2p.
Then there's DeRAM.
The simplest way I can describe it is "decentralized RAM."
It's focused on situations where multiple nodes need to read and update the same state. Anyone who's worked with distributed systems knows shared state can get messy pretty quickly, and DeRAM seems to be trying to make that process less painful and less slow.
That was the point where I started realizing Optimum isn't only thinking about how data moves. They're also thinking about what happens after the data arrives.
DeROM is a little different.
It's designed for data that's mostly read rather than constantly modified. Think things like caches, broadcast data, or information that's written once and then accessed repeatedly.
From what I've seen, DeROM is still on the roadmap, so it's not something people can use yet.
Looking at all three together changed how I think about the project.
At first I saw mump2p and thought, Okay, this is a data propagation project.
Now it feels more like they're trying to build a broader data infrastructure stack.
Not just moving data.
But sharing it.
Storing it.
Reading it.
And making it easier to use across distributed systems.
mump2p feels like the propagation layer.
DeRAM and DeROM feel more like memory layers that sit above it.
Right now mump2p is definitely the most mature piece of the puzzle. But if DeRAM and DeROM eventually come together with it, Optimum could end up looking a lot more like Web3 infrastructure than just another networking project.
Tomorrow I want to spend some time figuring out why DeRAM keeps emphasizing shared-state semantics.
That part still feels a little fuzzy to me :3
@get_optimum @cryptooflashh @blockchainjeff
@CryptoSundayz @aqccapita @shariaronchain
Hello everyone! gMum!
Today I spent some time learning more about the products that Optimum is building.
After reading so much about RLNC and data propagation, I originally thought Optimum was simply building a networking protocol.
But the more I looked into it, the more it seemed like the project has a much bigger vision.
Right now, Optimum is mainly focused on three products: mump2p, DeRAM, and DeROM.
At first, the names sounded pretty similar, so I was a little confused too 😂
But once I started looking at them one by one, their roles became much clearer.
The first product is mump2p.
This is where the RLNC technology I have been studying is actually being used.
Its goal is to help blocks, transactions, and other blockchain data spread across the network more quickly.
What I found interesting is that it works with existing libp2p infrastructure.
So rather than replacing everything, it seems to improve the efficiency of data propagation on top of systems that already exist.
It is also the only product that is currently available for builders and node operators to use today.
Because of that, a lot of the RLNC concepts I learned earlier connect directly to mump2p.
The second product is DeRAM.
The easiest way for me to think about it is as a decentralized version of RAM.
It is designed to make it easier for multiple nodes to read and write shared data with low latency.
For example, many distributed applications need several nodes to constantly access and update the same state.
DeRAM seems to focus on reducing the complexity and delays that can happen when managing that shared state.
The more I read, the more it felt like Optimum is not only thinking about how data moves, but also how data is shared after it arrives.
Then there is DeROM.
This one is a little different from DeRAM.
Instead of focusing on both reading and writing, it is optimized for data that is mostly read.
It is designed for situations where data is written once and then read many times, or where new information is continuously appended.
The examples I saw included things like caches and broadcast data.
From what I understand, DeROM is still on the roadmap and is not fully available yet.
At first, I saw mump2p and assumed Optimum was mainly a data propagation project.
But after looking at the full product lineup, it feels like the team is aiming for something broader.
Not just moving data faster.
But also helping networks share data, read data, store data, and use data more efficiently.
Mump2p feels like a propagation layer, while DeRAM and DeROM feel more like memory layers built on top of it.
Mump2p is the product that gets most of the attention because it already exists.
But if DeRAM and DeROM are successfully added in the future, Optimum may end up looking less like a networking project and more like a complete data infrastructure layer for Web3.
Tomorrow, I want to learn a little more about why DeRAM keeps emphasizing shared state and shared-state semantics :3
@get_optimum @cryptooflashh @blockchainjeff @CryptoSundayz @aqccapital @shariaronchain
Hello everyone! gMum!
Today I spent some time learning more about the products that Optimum is building.
After reading so much about RLNC and data propagation, I originally thought Optimum was simply building a networking protocol.
But the more I looked into it, the more it seemed like the project has a much bigger vision.
Right now, Optimum is mainly focused on three products: mump2p, DeRAM, and DeROM.
At first, the names sounded pretty similar, so I was a little confused too 😂
But once I started looking at them one by one, their roles became much clearer.
The first product is mump2p.
This is where the RLNC technology I have been studying is actually being used.
Its goal is to help blocks, transactions, and other blockchain data spread across the network more quickly.
What I found interesting is that it works with existing libp2p infrastructure.
So rather than replacing everything, it seems to improve the efficiency of data propagation on top of systems that already exist.
It is also the only product that is currently available for builders and node operators to use today.
Because of that, a lot of the RLNC concepts I learned earlier connect directly to mump2p.
The second product is DeRAM.
The easiest way for me to think about it is as a decentralized version of RAM.
It is designed to make it easier for multiple nodes to read and write shared data with low latency.
For example, many distributed applications need several nodes to constantly access and update the same state.
DeRAM seems to focus on reducing the complexity and delays that can happen when managing that shared state.
The more I read, the more it felt like Optimum is not only thinking about how data moves, but also how data is shared after it arrives.
Then there is DeROM.
This one is a little different from DeRAM.
Instead of focusing on both reading and writing, it is optimized for data that is mostly read.
It is designed for situations where data is written once and then read many times, or where new information is continuously appended.
The examples I saw included things like caches and broadcast data.
From what I understand, DeROM is still on the roadmap and is not fully available yet.
At first, I saw mump2p and assumed Optimum was mainly a data propagation project.
But after looking at the full product lineup, it feels like the team is aiming for something broader.
Not just moving data faster.
But also helping networks share data, read data, store data, and use data more efficiently.
Mump2p feels like a propagation layer, while DeRAM and DeROM feel more like memory layers built on top of it.
Mump2p is the product that gets most of the attention because it already exists.
But if DeRAM and DeROM are successfully added in the future, Optimum may end up looking less like a networking project and more like a complete data infrastructure layer for Web3.
Tomorrow, I want to learn a little more about why DeRAM keeps emphasizing shared state and shared-state semantics :3
@get_optimum @cryptooflashh @blockchainjeff @CryptoSundayz @aqccapital @shariaronchain
Hello everyone! gMum!
Yesterday, I learned how RLNC mixes and distributes data. Today, I spent some time looking into a concept that keeps coming up in RLNC: recoding.
At first, I thought it was just another round of encoding.
But the more I read, the more I realized that recoding plays a very important role in how RLNC works.
In a traditional network, intermediate nodes usually receive data and forward it to the next node almost unchanged.
They mainly act as messengers.
RLNC works a little differently.
A node can take the shards it has already received and create new shards from them before sending them on.
What I found interesting is that the node does not need to fully reconstruct the original message first.
It can generate new valid shards using only the encoded shards it already has.
Because of this, intermediate nodes become active participants in the propagation process instead of simply forwarding data.
The more I looked into it, the more this felt like one of the biggest differences between RLNC and traditional gossip networks.
Traditional gossip networks keep copying and forwarding the same data.
With recoding, RLNC keeps creating new pieces of useful information as the data moves through the network.
This allows information to spread through many different paths.
Even if some nodes or connections have problems, the overall propagation process can continue.
This also seems to be one of the reasons RLNC is known for strong loss recovery.
Even if some shards disappear along the way, other nodes can continue generating new shards from the data they already have.
As long as enough independent shards are collected, the original message can still be recovered.
That means the network can keep working even when some data is lost.
Another interesting point is that receivers are less likely to get the exact same data repeatedly.
Instead, they keep receiving newly mixed shards that provide fresh information.
Because of that, they can often collect the pieces needed to reconstruct the original message more quickly.
This can help reduce tail latency and ease network congestion.
The more I read, the more recoding felt like taking puzzle pieces, mixing them together, and creating brand-new puzzle pieces rather than simply making copies.
Because of that, Optimum does not seem to be focused only on sending data faster.
It looks more like an attempt to make the entire propagation process smarter and more resilient.
Tomorrow, I want to learn more about how this kind of system is actually implemented in real networks :3
@get_optimum @cryptooflashh @blockchainjeff
@CryptoSundayz @aqccapital @shariaronchain
@yjkang686510 @get_optimum @shariaronchain @blockchainjeff @tgogayi @aqccapital @cryptooflashh 헐!!!!!! 새우 그림고수엿자나?????????
Hello everyone! gMum!
Yesterday, I learned how RLNC mixes and distributes data. Today, I spent some time looking into a concept that keeps coming up in RLNC: recoding.
At first, I thought it was just another round of encoding.
But the more I read, the more I realized that recoding plays a very important role in how RLNC works.
In a traditional network, intermediate nodes usually receive data and forward it to the next node almost unchanged.
They mainly act as messengers.
RLNC works a little differently.
A node can take the shards it has already received and create new shards from them before sending them on.
What I found interesting is that the node does not need to fully reconstruct the original message first.
It can generate new valid shards using only the encoded shards it already has.
Because of this, intermediate nodes become active participants in the propagation process instead of simply forwarding data.
The more I looked into it, the more this felt like one of the biggest differences between RLNC and traditional gossip networks.
Traditional gossip networks keep copying and forwarding the same data.
With recoding, RLNC keeps creating new pieces of useful information as the data moves through the network.
This allows information to spread through many different paths.
Even if some nodes or connections have problems, the overall propagation process can continue.
This also seems to be one of the reasons RLNC is known for strong loss recovery.
Even if some shards disappear along the way, other nodes can continue generating new shards from the data they already have.
As long as enough independent shards are collected, the original message can still be recovered.
That means the network can keep working even when some data is lost.
Another interesting point is that receivers are less likely to get the exact same data repeatedly.
Instead, they keep receiving newly mixed shards that provide fresh information.
Because of that, they can often collect the pieces needed to reconstruct the original message more quickly.
This can help reduce tail latency and ease network congestion.
The more I read, the more recoding felt like taking puzzle pieces, mixing them together, and creating brand-new puzzle pieces rather than simply making copies.
Because of that, Optimum does not seem to be focused only on sending data faster.
It looks more like an attempt to make the entire propagation process smarter and more resilient.
Tomorrow, I want to learn more about how this kind of system is actually implemented in real networks :3
@get_optimum @cryptooflashh @blockchainjeff
@CryptoSundayz @aqccapital @shariaronchain
Hello everyone! gMum!! :3
Yesterday, I spent some time learning about RLNC, the core technology behind Optimum. Today, I decided to take a closer look at how it actually works.
At first!
I thought RLNC was simply a way of mixing data before sending it.
But the more I read, the more I started to understand why it works differently from traditional gossip networks.
The name RLNC still sounds complicated, but the basic idea is actually pretty simple.
In a traditional network, a node receives a message and then forwards almost the exact same message to other nodes.
Because of that, the same data gets copied and sent around the network again and again.
As the number of nodes grows, the amount of duplicated data also increases, which can make the network less efficient.
The RLNC works differently.
Instead of sending the original pieces of data, it breaks a message into smaller parts and mixes them together to create new encoded pieces.
Because of this, nodes do not need every original piece of data.
As long as they collect enough independent pieces, they can rebuild the original message.
I found this part especially interesting.
At first, I assumed mixing data would make everything more complicated.
But it actually helps the network use its resources more efficiently.
More information can be delivered using the same bandwidth, and even if some data is lost, there is still a good chance the original message can be recovered.
This seems especially useful in large distributed networks where connection quality is not always stable.
Optimum is applying this idea directly to data propagation.
In its mump2p protocol, data is shared as encoded shards, and intermediate nodes can even create new encoded shards from the ones they already received.
Another interesting thing is that nodes do not have to wait for the full message before forwarding data.
They can start passing information to other nodes as soon as they receive enough pieces.
Because of that, data can spread faster across the network, and packet loss becomes less of a problem.
The more I read, the more it felt like traditional gossip networks are passing around full copies of a document, while RLNC is more like sharing mixed puzzle pieces.
The receiver does not need every original piece. They only need enough useful pieces to rebuild the full picture.
Because of that, Optimum seems to be focused on more than just faster transmission.
It is trying to build a network that can continue delivering data reliably even when some information gets lost along the way.
Tomorrow, I want to learn more about why recoding is such an important part of this system!
@get_optimum @cryptooflashh
@blockchainjeff @CryptoSundayz
@aqccapital @shariaronchain
안녕하세요! gMum!
어제는 Optimum의 RLNC가 뭔지부터 조금 공부해봤는데 오늘은 그 부분을 조금 더 들여다봤습니다!
처음에는 그냥 데이터를 섞어서 보내는 기술 정도로 이해했어요.
근데 보다 보니까 왜 이 방식이 기존 gossip 구조랑 다르게 동작하는지 조금씩 보이더라고요.
RLNC라는 이름은 여전히 어렵지만 개념 자체는 생각보다 단순했습니다.
기존 네트워크에서는 메시지를 받으면 거의 그대로 복사해서 다른 노드에게 전달합니다 쓩~
그래서 같은 데이터가 네트워크 안에서 계속 반복되는 경우가 많아요.
노드가 많아질수록 중복도 늘어나고 생각보다 비효율도 커지게 됩니다ㅜㅡㅜ
반면 RLNC는 메시지를 여러 조각으로 나눈 뒤 그 조각들을 섞어서 새로운 형태로 전달합니다.
ㅂ
그래서 꼭 원본 조각을 전부 받을 필요가 없습니다.
충분한 수의 독립적인 조각만 모이면 원래 데이터를 다시 복원할 수 있다고 해요.
처음에는 데이터를 섞으면 오히려 복잡해지는 거 아닌가 싶었는데 오히려 반대더라고요.
같은 네트워크 자원 안에서도 더 많은 정보를 전달할 수 있고 일부 데이터가 사라져도 복구 가능성이 높아집니다.
특히 노드가 많고 네트워크 환경이 일정하지 않은 분산 시스템에서는 이런 차이가 더 크게 나타나는 것 같았어요.
Optimum에서는 이걸 실제 gossip 전파 방식에 적용하고 있습니다.
mump2p라는 프로토콜을 보면 데이터를 코드화된 샤드 형태로 전파하는데 중간 노드가 받은 샤드를 다시 재구성해서 전달할 수도 있다고 해요.
재밌는 건 노드가 전체 메시지를 다 받을 때까지 기다릴 필요가 없다는 점이었습니다!!!
일부만 받아도 바로 다음 노드에게 전달할 수 있으니까요.
그래서 전체 전파 속도도 빨라지고 패킷 손실이 발생해도 네트워크가 크게 흔들리지 않는다고 합니다.
계속 보다 보니까 기존 gossip은 문서를 통째로 복사해서 돌리는 느낌이고 RLNC는 퍼즐 조각을 계속 섞어서 전달하는 느낌에 더 가까운 것 같았습니다.
받는 사람은 원본 조각을 전부 받을 필요 없이 필요한 만큼만 모으면 되니까요.
그래서 Optimum가 이야기하는 건 단순히 더 빠른 전송이 아니라 손실이 발생해도 끝까지 데이터를 잘 전달하는 구조에 가까워 보였습니다.
내일은 재인코딩(recoding)이 왜 중요한지에 대해서도 조금 더 공부해보겠습니다!! :3
Hello everyone! gMum!! :3
Yesterday, I spent some time learning about RLNC, the core technology behind Optimum. Today, I decided to take a closer look at how it actually works.
At first!
I thought RLNC was simply a way of mixing data before sending it.
But the more I read, the more I started to understand why it works differently from traditional gossip networks.
The name RLNC still sounds complicated, but the basic idea is actually pretty simple.
In a traditional network, a node receives a message and then forwards almost the exact same message to other nodes.
Because of that, the same data gets copied and sent around the network again and again.
As the number of nodes grows, the amount of duplicated data also increases, which can make the network less efficient.
The RLNC works differently.
Instead of sending the original pieces of data, it breaks a message into smaller parts and mixes them together to create new encoded pieces.
Because of this, nodes do not need every original piece of data.
As long as they collect enough independent pieces, they can rebuild the original message.
I found this part especially interesting.
At first, I assumed mixing data would make everything more complicated.
But it actually helps the network use its resources more efficiently.
More information can be delivered using the same bandwidth, and even if some data is lost, there is still a good chance the original message can be recovered.
This seems especially useful in large distributed networks where connection quality is not always stable.
Optimum is applying this idea directly to data propagation.
In its mump2p protocol, data is shared as encoded shards, and intermediate nodes can even create new encoded shards from the ones they already received.
Another interesting thing is that nodes do not have to wait for the full message before forwarding data.
They can start passing information to other nodes as soon as they receive enough pieces.
Because of that, data can spread faster across the network, and packet loss becomes less of a problem.
The more I read, the more it felt like traditional gossip networks are passing around full copies of a document, while RLNC is more like sharing mixed puzzle pieces.
The receiver does not need every original piece. They only need enough useful pieces to rebuild the full picture.
Because of that, Optimum seems to be focused on more than just faster transmission.
It is trying to build a network that can continue delivering data reliably even when some information gets lost along the way.
Tomorrow, I want to learn more about why recoding is such an important part of this system!
@get_optimum @cryptooflashh
@blockchainjeff @CryptoSundayz
@aqccapital @shariaronchain
gmum !
Today I spent some time learning about Optimum(@get_optimum)
At first, I thought it was just another networking project, but the more I looked into it, the more interesting it became!
When people talk about blockchains, they usually focus on TPS or execution performance.
But even if a blockchain is very fast, it can still run into problems if data moves slowly across the network.
Blocks need to be shared, transactions need to be shared, and nowadays blob data also needs to move between nodes constantly.
Optimum seems to focus on making this data propagation process more efficient.
As I kept reading, it looked like the core technology behind it is RLNC.
The name sounded complicated at first, but the basic idea is actually pretty simple.
A message is split into smaller pieces, and those pieces are mixed and encoded before being sent across the network.
What I found interesting is that a node does not need to receive the entire message before forwarding it.
As soon as it receives enough pieces, it can start sending data to other nodes.
Because of this, information can spread across the network much faster.
Traditional gossip networks feel a bit like repeatedly copying and forwarding the same data, while RLNC seems like a more bandwidth-efficient way to distribute information.
One reason Optimum caught my attention is that the team is trying to turn these ideas into real products rather than keeping them at the research stage.
They are also building a protocol called mump2p, which is compatible with libp2p and aims to improve data propagation speed.
The project also talks about systems like DeRAM and DeROM, which makes it feel like more than just a networking solution.
The more I read, the more it seemed like Optimum is building a data infrastructure layer for Web3.
Validators need to receive data quickly, node operators want to reduce bandwidth costs, and dApps need a better user experience.
All of these things are connected to how efficiently data moves through a network.
Because of that, Optimum feels less like a project focused on creating a faster blockchain and more like a project trying to change the way data itself moves across blockchain networks.
Most people pay attention to the execution layer or the consensus layer, but sometimes the networking layer may be just as important.
@MurielMedard @kentlinyy
Most Popular Users
Elon Musk 
@elonmusk
240.2M followers
Barack Obama
@barackobama
119.3M followers
Donald J. Trump
@realdonaldtrump
111.6M followers
Cristiano Ronaldo
@cristiano
109.1M followers
Narendra Modi
@narendramodi
107M followers
Rihanna
@rihanna
97.3M followers
NASA
@nasa
92.1M followers
Justin Bieber
@justinbieber
90.6M followers
KATY PERRY
@katyperry
86.9M followers
Taylor Swift
@taylorswift13
80.7M followers
Lady Gaga
@ladygaga
72.3M followers
Kim Kardashian
@kimkardashian
69.4M followers
Virat Kohli
@imvkohli
68.7M followers
YouTube
@youtube
68.6M followers
Bill Gates
@billgates
63.5M followers
The Ellen Show
@theellenshow
62.5M followers
CNN
@cnn
61.9M followers
Neymar Jr
@neymarjr
61.2M followers
X
@x
60.9M followers
Selena Gomez
@selenagomez
60M followers