Recently, the Nexus @NexusLabs project has gained a lot of attention due to its airdrop interaction convenience of 'one-click mining' and the project's huge prospects, so I specially searched for information to learn about it, and here I share it with everyone. #Nexus #NEX #zkVM

Nexus claims to be a large-scale parallelized ZK (zero-knowledge) mining network, co-founded by Daniel Marin, Jens Groth, and others, all of whom have strong backgrounds in computer science and cryptography.

Since the project started in 2022, it has raised a total of 27.2 million USD, led by well-known VCs such as Pantera Capital and Dragonfly.

To understand the phrase 'large-scale parallelized zero-knowledge proof (ZK) mining network', one must know about ZK. Below is a brief summary of my understanding of ZK and why blockchain needs ZK.

📌 The Nexus project can be divided into two parts:

1. Nexus zkVM

zkVM can be simply understood as inputting any program or computation into zkVM, which will return a ZK proof process and result.

Developers do not need to learn cryptography or complex zero-knowledge development techniques, but the construction of zkVM itself is very troublesome. Therefore, the complex Nexus zkVM can help developers who require ZK proofs save a lot of time.

The main proof process is as follows (not exploring the specific ZK proof principles here, just discussing the proof generation process in Nexus zkVM):

1. Write -> Compile

Developers first write a regular Rust program, use a pre-compilation tool for processing, and then compile it into executable code targeting the Nexus Virtual Machine (NVM) instruction set using the compilation tool provided by Nexus (Nexus Compiler).

Note the distinction between NVM and Nexus zkVM

Nexus Compiler will compile high-level languages or specific DSLs (Domain-Specific Languages) into a concise and efficient low-level instruction sequence to maximize subsequent proof performance.

2. Execute and generate execution traces

After compilation, the program is executed in the NVM environment, and the entire execution process will be recorded as a series of 'execution traces'. This trace contains information about each step's state changes, memory accesses, etc., preparing for subsequent proof generation.

3. Chunking and distribution

NVM will split the entire execution trace into several chunks, and then distribute these chunks to different participating nodes through the orchestrator in the Nexus Network. This allows for parallel computation in a large-scale network, improving the speed and efficiency of proof generation.

4. Parallel proof and aggregation

Each node independently generates proofs (partial proofs) for the trace blocks it is responsible for.

These partial proofs will ultimately be aggregated by the orchestrator in the network into a unified, concise zero-knowledge proof, proving the correct execution of the original Rust program.

Due to the adoption of large-scale parallelism and recursive SNARKs, the final proof result is very small but can guarantee the correctness of the computational process.

Appendix: Memory consistency checking mechanism

Nexus zkVM uses a vector commitment method based on Merkle trees to track and verify the correctness of global memory read and write operations.

During multiple incremental verifiable computation (IVC) processes, this Merkle tree structure can ensure the correctness and consistency of each memory access, thus guaranteeing that the global node state does not undergo tampering or inconsistency.

2. Nexus Network

Nexus Network is the network that handles the 'execution trace blocks' and is composed of various computing nodes and orchestrators.

The orchestrator is mainly responsible for coordinating network resources, distributing block verification tasks to different nodes, verifying the work completed by nodes, and managing the current reward points.

Computing nodes need to run Nexus zkVM, execute tasks distributed by the computing orchestrator, submit results, and receive rewards through correct result feedback.

📌 Next, let's take a look at the current status of the project

A few days ago, the second testnet of Nexus was just released, attracting many participants. Currently, there are two ways to earn NEX points:

1. Run the web version, open the official website and click to start

2. Run the local version, install the node using the CLI command line

Currently, there are 30 node service providers offering computing on the testnet, including well-known DePIN projects like zklink and io.net.

🔴 To summarize

Compared to Succinct, it is found that Nexus is the ultimate version of Succinct, except that Succinct wants to take it slow and first run the ZKP network in the blockchain network before expanding to other industries.

However, this time Nexus goes straight to the point, with the ambition to create the world's zero-knowledge (ZK) proof generation network.

But as far as I know, ZK technology is currently mainly applied in the blockchain field. Nexus currently supports Rust and Solidity languages, which correspond to the Ethereum and Solana ecosystems. It can be said that Nexus is a strong competitor to Succinct.

Looking forward to what the overall development of the zkVM track will be like!