Optimum installs memory for Ethereum

The cryptocurrency industry is, to a large extent, a content creative industry. Countless heroes hope to attract attention with various creative ideas and complete the TGE of tokens. They want to exchange the big pie and U in your hands for my brand new Token. Even the Chaoshan Pony Ma would feel ashamed.

On April 15, Optimum completed a seed round of financing of US$11 million, led by 1kx and participated by well-known VCs such as Robot, CMT Digital and Spartan.

Besides the amount, Optimum also belongs to the increasingly rare blockchain infrastructure track. After the Move public chain, the market no longer buys into the concept of high-performance L1. The reincarnated Sonic (Fantom) and the newly entered Berachain are more like penny stocks around "complex" token economics.

The public chain is a foregone conclusion. What is even more difficult is that most public chains and L2 are cloud-customized products. If CEXs are all big AWS customers, then decentralized public chains may be on the same server.

Optimum is courageous, but do we really need decentralized high TPS products today?

Another form of Celestia

Optimum follows Celestia's lead in intercepting Ethereum L2, providing another possibility for virtualized storage and indexing layers.

Optimum has its own halo, which is not given by 1kx or other VCs. Its founder, Muriel Médard, is a professor at MIT. As one of the inventors of the Random Linear Network Coding (RLNC) technology, she has more than 20 years of research experience in distributed systems and basic computer theory.

In other words, Optimum has a starting point in the cryptocurrency industry second only to the wave of Turing Award winners in 2017, and RLNC has successfully leveraged tens of millions of dollars. Let us briefly introduce it.

RLNC was born in 2003, focusing ontransmitting data more securely and efficiently without changing the existing underlying Internet protocols. More concretely speaking, the transmission of network information is an inefficient process of continuous replication and quality compensation. When the network conditions of the two or more interacting parties are uncertain, the Internet tends to transmit the same information multiple times to reduce the "packet loss rate".

Specifically, there are only two ways to transmit information on the Internet. The first is to send large amounts of information and transmit it multiple times. Sooner or later, the information can be sent. It is always good to try more. The second is to wait for a clear reply from the recipient and extend the time to reduce data loss.

Either trade time for space, or trade space for time.

RLNC means both, saving both time and space, and building a new era of on-chain information transmission in a faster, better and more economical way.

RLNC's idea is to reduce the loss of information transmission by cutting the data into multiple data blocks or packets, and then encoding the data packets with constants, for example, a string of information is divided into data blocks 1, 2, 3, 4, 5, 6.

Each time we try to encode 1,2, we add a different random constant term to form

• • Packet 1 = random number 1 X data block 1 + random number 2 X data block 2

• • Packet 1 = random number 100 X data block 1 + random number 200 X data block 2

Data blocks can be mixed and configured to form multiple quadratic equations. In this way, any two receptions are sufficient to construct the original data packet. More data blocks can be configured multiple times.

Compared with traditional information transmission methods, RLNC can reduce the difficulty of information transmission to a constant level and greatly reduce transmission power consumption. Therefore, it is widely used in cloud computing, IoT, P2P and other scenarios.

So, what does this have to do with blockchain?

In Muriel Médard's view, blockchain is essentially a distributed system with multiple repeated backups, extremely difficult indexing, and extremely high time consumption.

Because any bill information needs to be recorded by all nodes, there are two ways to improve efficiency:

1. The first is to improve the throughput of nodes, that is, the Optimum P2P product, using RLNC to help each node improve its operating efficiency;

2. The second is to build a temporary index layer for existing systems such as Ethereum, temporarily store commonly used information and time-sensitive information, and use the computer's RAM concept to launch DeRAM products.

So, what does this have to do with Celestia?

Recalling the love-hate relationship between Celestia and Vitalik, Vitalik believes that only L2 that uses the Ethereum mainnet or EigenLayer as the DA layer can be called Rollup.

Ethereum provides security for L2/Rollup, and L2/Rollup pays for Ethereum's Blob space. The price is fair and reasonable, while Celestia's strategy is "I have a cheaper DA layer for sale here." Although everyone knows that it is not as safe as Ethereum, it wins in terms of quantity. Regarding the concept of Celestia and DA layer, you can refer to the previous article The war on Ethereum Rollups (STARKNET) has ended, and the new narrative DA is ready to go.

The idea of ​​Optimum is the same. Through RLNC technology, it can provide cheaper access and its security is guaranteed by mathematical mechanisms.

Ethereum is lacking in core and soul

From a technical point of view, Optimum can indeed reduce node latency to enhance node profitability, but in reality, the centralization of Ethereum is basically a foregone conclusion, for no other reason than higher efficiency.

In terms of specific implementation, Optimum P2P has been promoted in advance, while DeRAM is still under development. In Optimum's view, blockchain is in a typical "lack of core and soul" semi-crippled state.

In terms of modern computer system architecture, the input and output, CPU, operating system, and storage functions constitute the von Neumann architecture.

Optimum believes that the current blockchain has no memory and can only use the operating system to directly call the data of the hard disk (node). The key is still the 3.5-inch floppy disk, so it is necessary to add RAM (memory) function (RAM is not the same as the concept of memory, which is simplified here).

Of course, EVM or Ethereum, as the core of the entire public chain system, is cumbersome and is a problem that is difficult to completely solve, just like the DOS system cannot run 4090. Optimum can theoretically be integrated with any L1/L2, that is, Scale Any Chain.

In fact, do not increase entities unless necessary. The expansion and speedup of public chains have reached critical values. It is an indisputable fact that most public chains and L2 architectures are in cloud services. From the perspective of incentivizing nodes to join the Optimum network, this is more like a new project competing for the trust of large investors, first raising TVL, and then choosing the right time to ship to retail investors.

Conclusion

After seeing Optimum get financing, your first reaction is the rebirth of VC coins?

From the financing data of Decentralised.co, the financing in the first half of April is almost catching up with the entire amount in March, especially a16z CSX Optimum has also officially joined the fray. Although I am not optimistic about its integration prospects with Ethereum, it is still worth our attention in terms of expected tokens and technological exploration. Again, Sonic and Berachain actually have little to do with technological progress. Some new ideas are always good. Think about what other elements of Ethereum can be disassembled and tokenized, and leave it to the founders to think about.