A truly decentralized Extensible Blockchain

Today, the blockchain revolution is wrestling with the Internet, which was born in the 1990s.

AMBG engineers have devoted thousands of hours to blockchain research and development, with the ultimate goal of replacing the existing economy in a safer and decentralized way.

However, despite the great potential of blockchain technology, Bitcoin, Ethernet Square and most other traditional distributed blockchain networks are still plagued by the fundamental problem of expansion.

The problem of blockchain expansion determines whether this brand-new technological world can really become a reality, but it does have a technical bottleneck that is difficult to break through, including the limitations of technology itself, block size, response time, cost and so on.

The expansion of Bitcoin has something to do with the limitations of its own technology. When processing a new transaction, each node will write down the corresponding information in the ledger.

In this way, with the increase of payment history information, the block chain volume also increases, resulting in the normal operation of small nodes. Another expansion problem is the volume of the block. Initially, each block in the bitcoin chunk chain had a capacity of 1MB, and each block could store about 2020 transactions.

With the increase in the popularity of bitcoin, the number of bitcoin transactions has doubled in recent years, causing the bitcoin network to take 10 minutes to confirm a block, and it usually takes a long time to verify an unconfirmed transaction. In addition, due to the increasing computing power required for mining, transaction costs have naturally gone up.

Currently, many new blockchain projects are trying to improve transaction processing speed (throughput), but new solutions such as Rootchain in EOS and dPoS,Quarkchain in TRON have to sacrifice some key elements, such as decentralization and security, to significantly improve performance. Although such a system runs very quickly, it can only be regarded as a semi-centralized system, which loses the core idea of block chain-decentralization.

As a solution for block chain expansion, fragmentation can significantly improve network performance without losing security and decentralization.

AMBG solves the problem of block chain expansion by introducing state slicing into the block chain. Because each node only needs to run and store part of the block chain data to complete the transaction, the transaction processing workload is apportioned, which greatly improves the scalability of the block itself.

What is status fragmentation?

Suppose there are three nodes A, B, C and a data T to be verified. Compared with A, B, C, each node needs to store and verify the whole data T, and the status fragment divides the data T into three parts: T1 T2 and T3.

After that, A _ memory B and C will store and validate a part respectively.

It is not difficult to imagine that the efficiency of such a network will increase exponentially.

A simpler explanation is that if the United States is divided into different states, although each state (that is, sliced) is part of the United States (AMBG network), they all have their own specific laws, borders and population subsets, but share a language and culture, and are part of the country as a whole.

Take Ethereum as an example, all nodes in the Ethereum network store all the states of the block chain, including account balances, data and intelligent contract codes.

However, with the increasing scale of the network, the cost of gas is getting higher and higher, and the transaction confirmation time is getting longer and longer, which greatly reduces the practical value of ethernet.

The upper limit of the performance of Ethereum is the processing speed of a computer.

The performance of Ethereum can not increase with the increase of the number of nodes, on the contrary, it even decreases slightly.

Sharding provides a shortcut to break through this technical bottleneck: sub-nodes are divided into slices, and each shard handles different transactions separately, so that the system can handle many transactions at the same time, and the throughput is naturally significantly improved.

Zilliqa adopts the scheme of network fragmentation, which allows the network to be divided into smaller node groups, each of which can be called slicing. To put it simply, suppose a network with 1,000 nodes divides the network into 10 shards, each shard consists of 100 nodes, and if each shard can process 10 transactions per second, then all the shards together can process 100 transactions per second.

But in Zilliqa, each node must save the state of the whole block chain in order to deal with the transaction successfully. when the volume of the block chain increases gradually, it will become more and more difficult to keep a complete account book, the requirements for node equipment will increase rapidly, and the transaction cost will also increase. As a result, the state shards used by AMBG undoubtedly provide a better option-each shard saves a portion of the chunk state.

How AMBG avoids 1% attacks:

Although the sharding technology significantly reduces the transaction completion time and increases the transaction processing volume, at this stage, the biggest obstacle to the landing of the sharding technology is still its potential security problem, that is, the “1% attack”.

In blockchains that use PoW, such as Bitcoin, a 51% attack is possible when the attacker has most of the hash computing power of the network.

Once it happens, the attacker can “double flowers”, ask for all rewards, block the transaction, and so on.

However, it takes a lot of power and equipment to have 51% or more of the hash computing power of the entire network. Currently, the cost of a 51 percent attack on a Bitcoin network is $8 billion and the daily electricity bill is $12.8 million.

Suppose a blockchain network is divided into 100 pieces, that is, 100 fragments, each with 1% hash computing power. Then, in essence, the attacker only needs to concentrate his hash computing on a single fragment to achieve control over the fragment, which will undoubtedly affect the security of the whole network.

The same problem applies to PoS systems.

An effective way to prevent such attacks is to prevent attackers from concentrating their hash or token rights on a particular piece. By slicing and randomly sampling the rights and interests of PoS tokens, AMBG can effectively eliminate the rights and interests of attackers in a certain fragment, thus eliminating the possibility of 1% attack.

AMBG’s PoS mechanism ensures that:

-the attacker cannot select the shards he wants to join.

-attackers can’t know in advance which part they will be assigned to.

If the cards are reshuffled every once in a while, how long will it take for a node to download the state of the shards?

When a verifier adds a new shard, they need to quickly synchronize the shard data to validate the new transaction.

In the traditional process of downloading block history, it takes many days to completely synchronize all historical information, but in AMBG’s fast synchronization algorithm, nodes only need to download the block head of this fragment and the block data of the previous cycle to start verifying new transactions. Therefore, nodes will be quickly prepared in the process of reassignment by verifiers.

Cross-chip Communication of AMBG

When transactions are assigned to different fragments, efficient communication between fragments becomes crucial. If the communication between slices is difficult, like isolated islands, then the whole slicing system will lose its meaning.

For example, suppose you want to book a train ticket and a hotel in a shard protocol network, and the train ticket transaction is in one segment, and the hotel transaction is in another. For users, the result you want is the success of booking both the ticket and the hotel at the same time, not one success and the other failure.

AMBG supports cross-shard trading. Shards can not only communicate with each other, but also perform shard operations synchronously. When each new block is created, the block chain header is sent to the beacon chain through an Kademlia routing protocol.

In this way, each node in the AMBG network keeps a routing table containing the distances of all nodes.

When a message from shard An is to be sent to shard B, the node in shard A will consult the routing table and send the message to the nearest node. This method ensures that the information is only transmitted through the shortest path and reaches the target shard in the fastest way. Compared with the common “Gossip broadcast” used by projects such as Zilliqa and Hashgraph, the Kademlia routing protocol reduces the overall network load in terms of the underlying mechanism.

For example, in the Kademila setting, in a network with 10,000,000 nodes, communication between any child nodes requires a maximum of about 20 propagated drop point (hops).

The Hamony protocol adopts the technological breakthrough that has been verified in research and practice, the security node allocation and state slicing technology based on PoS, which not only makes the whole block chain system more secure, but also realizes the real decentralization.