Improvement and optimization

An integral reason for the growth of the DIGITAL - HI TECH Series (DHT) was the increased integration with existing enterprise systems. Tight fusion of existing main chain modules (Yao, Andrew Chi-Chih "Protocols for Secure Computing." In FOCS, vol. 82, pp. 160-164 In 1982.

Bünz, Benedikt, Jonathan Bootle, Dan Boneh, Andrew Poelstra, Pieter Wuille and Greg Maxwell. "Bulletproof: Concise Proof of Confidential Transactions and More." In 2018, IEEE Symposium on Security and Privacy (SP), pp. 315-334. IEEE, 2018) blockchains impede system integration. In addition, many main chains with high TPS do not meet the performance requirements for transmission of high concurrent requests.

Structural optimization: DIGITAL - HI TECH (DHT) modularizes the development of databases, networks, consensus, and other technical features, before storing them separately in containers. A module call can be made through the middleware, resulting in the separation of each module. Furthermore, the infrastructure layer can automatically perform a wide range of operations such as registration, activation, and container upgrades using middleware.

For subchain developers, the code of the DIGITAL - HI TECH (DHT) Chain infrastructure layer boasts clear logic and allows for easy rebuilding. In addition, DIGITAL - HI TECH (DHT) created a new routing and messaging mechanism that completely decoupled the network and consensus layers, increasing the scope of consensus beyond PBFT, thus facilitating independent development. consensus at the subchain level.

DPOS + PBFT consensus optimization DIGITAL - HI TECH (DHT) Chain uses DBFT, a new consensus mechanism that is a merger of DPOS and PBFT consensus mechanisms. In the DBFT consensus mechanism, the proper functioning of the blockchain depends on the trustees, who are fully equivalent. Responsibilities of the trustee include:

• provide stable servers to ensure the nodes work properly, ensure the nodes will collect network transactions

• ensure nodes will collect network transactions

• ensure nodes will validate transactions and package transactions into blocks

• make sure nodes broadcast blocks and add blocks to their own database after other nodes verify

• Participate in the administration and maintenance of the DIGITAL - HI TECH Chain (DHT)

The trustee's node will receive block rewards and transaction fees for performing their tasks.

Seven trustees will be selected to help run the DIGITAL - HI TECH (DHT) 1.0 Darwin Chain Test Network. Any holder of the DIGITAL - HI TECH (DHT) token can participate in both the trustee voting and voting process. The influence of each token holder on the voting results is proportional to their token holdings. Votes may be exercised and withdrawn at any time prior to the conclusion of the trustee election. The seven users with the highest number of votes at the end of each election will become trustees of the DIGITAL - HI TECH (DHT) Chain. In return for the responsibility of creating blocks and maintaining the system, the seven elected commissioners will be duly rewarded.

The basic purpose of the election is based on user voting, selecting seven highly committed users who can strongly support the development and operation of DHT. They are expected to operate the nodes to ensure high system performance and efficiency and also contribute to the development of the project to the best of their ability.

• The specific implementation of the election is outlined below:

dlist_i = get N delegates sorted by votes loop dlist_i = shuffle(dlist_i) slot = global_time_offset / block_interval pos = slot %N if dlist_i[pos] exists in this createBlock(key pair of dlist_i[pos]) else ignored.

• Voting process:

DHT's DBFT is a byzantine tolerant state machine clone and is implemented with a three-stage voting process: "Prepare", "Prepare", and "Commit".

Stage 1: Prepare in advance

The team leader is responsible for receiving customer requests

The Head will send a signed "Prepare" message (see, seq) to the other Validator over the network.

Stage 2: Prepare

Each Validator must decide whether to accept the Leader's proposal after receiving the "Prepare" message. If the Validator accepts the proposal, they will send their signed "Prepare" message to all other champions, otherwise they will not transmit any messages.

Validators who sent a "Prepare" message will initiate a "Prepare" Phase.

Heroes that receive more than three "Ready" messages will go into "Ready". This set of "Preparation" messages is collectively known as the "Preparation Certificate".

Stage 3: Commitment

If a "Prepare" champion decides to commit, he will send a signed "Commit" message to all champions, otherwise he will not transmit any messages.

The champion sending the "Commit" message will initiate the "Commit" Phase.

If generals receive more than three "Commitment" messages, they will follow the instructions of the message. This also means that the proposal has reached consensus.

After executing the instructions of the message, the champion enters the "Committed" state and reports the execution (Reply) to the byzantine client.

Once the answer has been submitted, the current process will end and the champion will wait for the next suggestion.

Advanced substring function

Data recovery is required if there is a substring data error during the synchronization of substring data with the main chain., Although the initial level DB does not support this operation, nesting and substring recovery is done by DIGITAL - HI TECH (DHT) self-development can solve the problem.

Staking Mechanism DIGITAL - HI TECH (DHT)

DIGITAL - HI TECH (DHT) token holders who wish to participate in the verification of transactions over the DIGITAL - HI TECH (DHT) network can contribute their tokens to support nodes becoming candidate nodes. Staking participants will be rewarded based on the staking time and the number of tokens staking. When a participant needs tokens for other transactions, he can submit a withdrawal request and will receive his staked tokens along with staking rewards in his wallet after the current DBFT cycle.

By actively participating in the network through staking, token holders enhance cybersecurity, protect their rights as token holders, and receive staking rewards. Without actively staking, the tokens held do not grant the holder any rights and benefits from staking. Thus, the staking economy is an important trend for cryptocurrencies and is inseparable from every token holder.

Smart Channel DIGITAL - HI TECH (DHT)

An independent data cluster called a Smart Channel may need to be established between nodes in an DIGITAL - HI TECH (DHT) network due to data isolation or other special requirements. One of the nodes will broadcast a transaction to create a Smart Channel while the other nodes will apply to participate in the generated data cluster. The influence of each node in this cluster is equal. Nodes can communicate privately, process transactions independently, and jointly manage all copies of the ledger associated with the cluster. It can be seen that the cluster has a structure similar to that of a circle of friends, although the members of the group are not necessarily friends. A person can have multiple friend circles, and each friend circle has specific activities they want to do. The different friend circles may be completely independent or may overlap, but each friend circle is itself an entity and has its own "rules".

Cooperation between businesses or organizations is often closely linked and can overlap. Assume that there are four organizations named Org1, Org2, Org3 and Org4, where Org1, Org2 and Org3 together form a financial chain alliance, while Org2 and Org4 form a supply chain alliance. DHT's smart channel allows them to build multiple clusters based on specific business needs, create different consortium chains, and add new chains in real time as needed. By supporting clusters, DIGITAL - HI TECH (DHT) allows flexible construction of conjugate chains according to this architecture.

Virtual Machine Optimization:

DIGITAL - HI TECH (DHT) has improved the existing TVM to meet the business needs of the DIGITAL - HI TECH (DHT) series:

• Added the ability to execute smart contracts on different chains

• Increase orders related to reputation

• Upgraded gas prices, allowing automatic adjustment to the configuration and demand on each subchain

• Redesigned infrastructure database

Database Optimization

High concurrency requests put a lot of pressure on mechanisms like caching. This means that blockchains that boast of their 'high TPS' capabilities may not necessarily be able to handle high transaction volumes as sold in the market. Unlike these blockchains, DIGITAL - HI TECH (DHT) optimizes necessary mechanisms like caching, databases, etc. and uses granular keys inside the lRU cache to ensure high performance even when TPS reaches 5,000.

The DIGITAL - HI TECH On-chain Level DB (DHT) database uses Hash Patricia Trie15 (Hitton Prefix Tree, hereinafter HPT) technology to store user account state and changes in account state.

HPT is a multi-branch tree data structure. Each node in the tree consists of 4 properties: Sequence, Value, Children and IsLeaf. The HPT root node's Value property stores the state hash of the current database for block verification. The Sequence property is the only way to get a specific Complete Key.

When a user's account information changes, the database makes corresponding modifications to the HPT to reflect the changes in the current database state. Based on the hexadecimal string key of the user account, the database will conduct a thorough search starting from the root node until a certain leaf node is found. The Sequence attributes of all nodes on the search path are then sequentially concatenated to obtain a complete key.

Two main advantages of HPT:

• First, the irreversible and extremely low collision properties of the tree structure and hashing algorithm improve the convenience and reliability of database states.

• Second, the database's critical value design and prefix tree's data compression greatly improve the efficiency of querying and modifying the database state while reducing computational costs. maths. The database can be quickly updated to reflect changes to the user's account information by operating the same number of tree nodes as the depth of the HPT. When making multiple modifications to the account information in a transaction, the increase in computation required to modify the HPT is much less than the increase in the amount of account information. The larger the amount of account information, the more significant the MPT calculation.

Compared to the common prefix tree structure used by BSC to store account information, DHT's HPT unifies and optimizes the design of tree nodes instead of applying the method of dividing tree nodes into empty, leaf nodes. , extension node and branch node. DHT's HPT reduces prefix tree height and thorough search time, thereby improving modifiability and query ability.

(Kniesburges, Sebastian and Christian Scheideler. "Hashed Patricia Trie: Efficient Longest Prefix Matching in Peer-to-Peer Systems." In the International Conference on Algorithms and Computation, pp. 170-181. Springer, Berlin, Heidelberg, 2011.)