Q: Does the system have a corresponding Fault Tolerance Mechanism in view of Byzantine node or incorrect nodes?
A: Yes, it supports PBFT (Practical Byzantine Fault Tolerant), the system will automatically right down in view of incorrect nodes, and the system with three or more nodes can run smoothly.
Q: Does the system have a corresponding handling mechanism in view of the simultaneous occurrence of the transaction?
A: Yes, double-spend problem is the key consideration, solved with multiple verification, inverse operation and other handling mechanisms. If there are crossovers in the synchronous transaction or the transaction is chained, by means of asynchronous reverse sorting, big data will be executed in order. If the system is irretrievable, the rollback will be triggered and released.
Q: In the case where the number of nodes is relatively low, is it possible to effectively control the transaction duration to avoid the situation that subsequent transactions cannot be verified in time?
A: Yes, we support the duration controlling. The duration is determined by real-time available resources. If it times out, it will be released and reversed into consensus rollback, terminating and releasing the resources.
Q: In view of asynchronous communication mechanism, if data encounters communication failure, etc., will there be data inconsistency in different nodes?
A: The start-up synchronizes with high speed on basis of size partition and following sharding, tolerating fault, and the system has a memory rollback function, catastrophic rollback, if the excitation index is not enough, it will automatically right down.
Q: In view of special cases such as server host failure, will there be data consistency or security problems caused by memory data loss?
A: Data storage has a dedicated algorithm storage module to break up and reorganize physical data.
A: Follow the instructions for the TESTING button on the website.
Technical development can only be implemented after the OBT is completed, and the major network is online. To go through the normal process to graft to the underlying application, the data generated during the OBT will be invalid.
A: MAC does not collect limited supernodes.
Specific transactions are collected intelligently, dynamically, and merit-based, according to the resources and calculation processing capacity. The so-called faster theory means that by reducing the consensus, it should be treated differently with MAC.
The MAC algorithm includes split reorganization and the forward/reverse execution algorithm. Through a small scope test, the more nodes, the more dynamic proxy resources will be, and the faster the efficiency.
A: For industries that need the decentralized technical solutions of blockchain as well as high TPS requirements. Examples include:
• Cross-border e-commerce
• Foreign exchange
Additionally, MAC is an easy-to-use system for developers, so it can be applied to different industries in the development stage of applications.
A: Thousands of transactions per second for the whole network may be sufficient for private chains or even consortium chains, but for a public chain, demand is far from being met.
Any public chain project using the single-chain DPOS structure cannot meet the needs for future expansion. Using DPOS as a base, MAC adopts NDPOS to meet the strong data communication of quasi-real-time cross-sharing.
Unlike the eventual consistency of the unpredictable transaction confirmation time of the DAG structure, NDPOS provides strong consistency across inter-chain transactions through the peer-to-peer, multi-activity mechanism of the blockchain.
At the same time, NDPOS realizes the infinite expansion of the entire network sharing numbers through the mechanism of hierarchical proxy, which solves the problems of the single-chain ledger.