Exploring Sharding Technology: The Innovative Journey from Ethereum to Shardeum

On September 15, 2022, Ethereum completed the highly anticipated merge (Merge). This upgrade transitioned Ethereum from a Proof of Work (PoW) mechanism to a Proof of Stake (PoS) mechanism, but did not directly bring higher scalability, security, and sustainability. Achieving these goals requires more comprehensive solutions, including a mainnet with Sharding capabilities and Layer2 solutions that enhance scalability.

Sharding is a scalability solution under the scalability trilemma. It divides the nodes in the network into smaller groups, handling different sets of transactions and achieving parallel processing. By sharing the burden of processing the entire network's data, sharding can improve processing efficiency, similar to a store adding checkout lanes to reduce waiting time.

Although the principle of Sharding is simple and straightforward, there are still many challenges in the implementation process. This article will outline the development direction and dilemmas of Sharding technology, creating a roadmap for Sharding explorers. At the same time, by comparing existing Sharding solutions, it aims to identify common issues and propose a new exploration direction: Shardeum and dynamic Sharding.

About "Sharding"

Starting from Ethereum as the origin point, we can categorize blockchain scalability methods into two main types:

Vertical Scaling(: Achieved by enhancing the performance of the existing hardware in the system. This method is simple and effective, suitable for latency-sensitive scenarios, but it will limit the degree of decentralization of the network.

Horizontal Scaling): There are mainly three approaches:

1. Distributing the transaction computation load across multiple independent blockchains.
2. Modular blockchain, dividing the infrastructure into execution layer, data availability layer, and consensus layer.
3. Split a blockchain into multiple shards for parallel execution.

These scaling solutions do not exist in isolation; each solution seeks balance within the impossible triangle, designed with economic incentive mechanisms to achieve effective balance at both macro and micro levels.

To discuss "Sharding", we need to start from scratch and sort out a few key issues:

1. How to determine which shard a full network node/validator belongs to? That is, network sharding (Network Sharding).
2. How to determine which shard each transaction is allocated to? That is, transaction sharding (Transaction Sharding)
3. How is blockchain data stored across different Sharding? That is, state sharding (State Sharding)
4. How to avoid the fragmentation of the overall system security?

( Network Sharding )

Network sharding refers to dividing the blockchain network into shards, allowing each shard to process on-chain transactions and compete for the right to maintain the ledger. The main challenge is that as nodes are assigned to different shards, the difficulty and cost for attackers are significantly reduced.

The solution is usually to build a certain degree of unbiased randomness to minimize the attacker's success probability. For example, Ethereum randomly selects a validator for a certain Sharding every 6.4 minutes and rotates them.

However, the randomness in blockchain itself is a challenging topic. Many existing designs have developed separate chains to maintain the entire network, such as Ethereum and Near's Beacon chain, PolkaDot's Relay chain, and Cosmos's Cosmos Hub.

Transaction Sharding (

Transaction sharding refers to the establishment of rules to determine how transactions are allocated to various shards. The ledger model of the blockchain will influence the development of transaction sharding.

Currently, there are two main accounting methods:

UTXO Model: A typical representative is Bitcoin. Under the UTXO model, transaction sharding requires cross-shard communication. One possible way is to use hash functions to distribute transactions across different shards.

Account/Balances Model: Represents something like Ethereum. As long as transactions are sharded by the sender's address, multiple transactions from the same account can be processed in the same shard, effectively preventing double spending. Therefore, most blockchains that utilize sharding technology adopt an account ledger system.

) State Sharding ###

State Sharding refers to how blockchain data is allocated and stored across different shards. This is the biggest challenge of sharding technology, because under the sharding mechanism, transactions are allocated to different shards based on their addresses, and the state is only stored in the shard where its address is located.

The main challenge is handling cross-sharding ( Cross-Sharding ) transactions. For example, when account A transfers to account B, and the two accounts are located in different shards. In this case, cross-shard communication is required, which may reduce transaction processing efficiency.

There are two main approaches to solving this problem:

Synchronous Sharding ###: Also known as Tight Coupling (. Whenever cross-shard transactions are executed, the related blocks occur simultaneously, and the nodes of each shard collaborate to execute the transactions.

Asynchronous Sharding ): Also known as Loosely Coupling (. This method is more widely used, such as in NEAR, Ethereum, Cosmos, Kadena, etc. The biggest challenge is to ensure atomicity of transactions.

![Detailed Explanation of the New Public Chain Shardeum: Another Possibility of Sharding])https://img-cdn.gateio.im/webp-social/moments-6e8d3331d7d68cb512eb2eb47bd9064d.webp(

Exploration and Attempts of Sharding

We will review the sharding technology solutions of several well-known public chains to explore the innovativeness of Shardeum.

) Calculate Sharding

Zilliqa is one of the earliest smart contract platforms to attempt Sharding. It uses a computational sharding approach to divide the work of validating transactions across different shards. However, since each node still receives all transactions and updates all account states, network bandwidth and storage operations remain bottlenecks.

( Static Sharding

A more generalized sharding method is to divide the account address space into fixed-size shards and allocate nodes to different shards. Platforms such as Near, Elrond, and Harmony adopt this method.

![A Comprehensive Explanation of the New Public Chain Shardeum: Another Possibility of Sharding])https://img-cdn.gateio.im/webp-social/moments-4227a2e49f76cd01b23d7b5398e51a3c.webp(

)# The concept of data sharding in Ethereum

Ethereum plans to implement data sharding to increase availability. One approach is sharding based on "data availability," using "data availability sampling" to confirm whether all data is available. Another approach is to add extra features to each shard, making it more like the current Ethereum mainnet.

(# Harmony

Harmony adopts a standard Sharding method, with a beacon chain coordinating multiple shards. It uses an effective Proof of Stake ) EPoS ( mechanism and secure random sharding technology to distribute the tokens staked by large holders across multiple shards, preventing a single shard from being attacked.

)# Elrond

Elrond employs adaptive state sharding, dynamically adjusting the number of shards. It uses a secure proof-of-stake consensus mechanism and achieves fast cross-shard transaction finality through the meta chain ###.

Near

Near proposed the Nightshade protocol (, modeling the system as a single blockchain with sharding at the block level. It does not have sharding chains; instead, all block producers and validators are building a main chain.

![Comprehensive Explanation of the New Public Chain Shardeum: Another Possibility of Sharding])https://img-cdn.gateio.im/webp-social/moments-21183f66edba6389ed99a8c668bb88d6.webp###

Shardeum and Dynamic State Sharding

Shardeum proposes two novel solutions to address the limitations of existing Sharding technology:

Transaction-level Consensus

Shardeum combines Proof of Quantity ( PoQ ) and Proof of Stake ### PoS ( to achieve consensus at the transaction level instead of the block level. This allows for simultaneous processing of transactions across shards, eliminating the complexity required to ensure atomic processing.

) Linear Expansion

Shardeum uses dynamic state sharding, allowing each node to hold different and overlapping address ranges. This approach is more complex but achieves true linear scalability.

Shardeum currently has 10 Sharding, with each Sharding having 128 nodes, totaling 1280 nodes. The mainnet is expected to launch in the fourth quarter of 2022. On the testnet Liberty 2.0, Shardeum has demonstrated a capability of 100 TPS.

![Detailed Explanation of the New Public Chain Shardeum: Another Possibility of Sharding]###https://img-cdn.gateio.im/webp-social/moments-678011ec0d792ac05e80a6b209c82d43.webp(

Conclusion

A truly sharding and scalable blockchain needs to be built from scratch, and similarly, a strong community requires continuous effort. After the launch of Shardeum's testnet, it gained widespread attention, although it also faced skepticism, we believe that the path to progress is always winding.

As the ancient saying goes, "A country has six roles, and the hundred crafts reside in one of them. Some sit and discuss the Dao, while others act on it." Jsquare hopes that more builders will join the exploration of decentralization and Sharding. We believe that only when Web3 achieves mass adoption can high-quality, high-performance decentralized public chains stand out. Those communities that have undergone questioning, discussions, and repeated demonstrations can go further on the path of decentralization.

![In-depth Explanation of the New Public Chain Shardeum: Another Possibility of Sharding])https://img-cdn.gateio.im/webp-social/moments-d20b7d1dbc15c9f665151aba3a8d52de.webp(

![Detailed Explanation of the New Public Chain Shardeum: Another Possibility of Sharding])https://img-cdn.gateio.im/webp-social/moments-93daeddd2b8a13824f6237ed7fd61163.webp###