3.1.3 Unique contribution
DGT is a distributed ledger technology that allows organizing interactions between organizations (Businesses) within the framework of a consortium-based approach. Several properties derived from the F-BFT consensus make it possible to position DGT as a platform for ecosystems, which can implement complex B2B2C and B2B2B models. Consortium-based networks, including DGT, are suitable for creating hybrid business communities. Some of DGT’s features include:
The storage system is based on DAG, similar to IOTA, Hedera, Hashgraph, Orumesh, DagCoin, Byteball, and Nano. The DGT approach differs from those systems in that it allows voting in a federated network structure with a customizable topology.
The federated approach to voting is actively used by solutions such as Ripple, and Stellar. DGT applies this ideology onto a horizontally scalable DAG and, through the rotation of leaders in clusters, provides a dynamic topology. This resolves the problem of the otherwise apparent compromise of network safety versus liveness.
The consortium-based consensus system allows to maintain flexibility, without losing speed and interoperability. For example, Hyperledger Fabric is aimed at private peer-to-peer networks and requires the formation of special support structures (sidechains). The ICON solution uses a special Loopchain Fault Tolerance mechanism for interacting with . Unlike these solutions, DGT implements a dynamic topology on top of DAG, allowing for a high degree of asynchrony in the network.
Having some of DGT’s functional properties intersect other existing solutions allows for an edge computing environment to be implemented, particularly through the simultaneous use of several well-known approaches combined within the framework of the platform. The solutions close to DGT in architecture, such as , do not have a clear focus on ecosystem components:
The horizontally scalable DAG storage architecture combined with federated networking.
Dynamic network organization due to topology processor.
A single ontological layer due to the support of several families of transactions.
The implementation of the F-BFT consensus, based on the probabilistic ratio of the time the network takes to vote and the polynomial interpolation of the voting results, allows for an effective and unique technical solution with broad functional properties and exceptional interoperability. This is necessary for constructing business systems that achieve a required level of .
DGT was initially built on a well-known Hyperledger Sawtooth framework (see DGT. Technology Roadmap, 2021 | Available on corporate site), led by the Intel Corporation. The philosophy, architecture, and implementation of Sawtooth most closely represent the goals and objectives of DGT. DGT inherited some low-level technical features, while also adding several of its own – as presented in the table below:
| FEATURES | Intel Sawtooth | DGT Network |
|---|---|---|
1. General architecture | DGT supports the underlying Sawtooth architecture in terms of modularity, distribution of core data handlers, event streams, and so on. | |
2. Low-level protocols | DGT mainly inherits Sawtooth solutions, including the use of ports, the ZeroMQ asynchronous library, virtualization, and flow management. | |
3. Cryptography | DGT redesigned the cryptographic system and can support several alternative systems, including the ECDSA curve secp256k1, as well as OpenSSL. | |
4. Type of DLT | Private/Public | Consortium-Based/Hybrid |
5. Network solution | One-level; the GOSSIP protocol is used for node discovery and transaction propagation | In a federated network; the interaction between nodes is determined by the topology processor (one of the transaction families) |
6. Consensus | Even though the framework supports a reloadable consensus mechanism, the leading algorithm is PoET, a consensus based on Intel chips. At the end of 2019, Intel announced support for the PBFT consensus. | The F-BFT algorithm is based on reaching consensus in clusters (federation) and then disseminating validated data to the entire network. |
7. Data storage | In blocks, physical storage is organized in reloadable data stores (LMDB, OpenTS, and other NoSQL databases). Sawtooth uses a Merkle-Radix tree for storing data for transaction . | DAG, is a directed acyclic graph, whose vertices contain batches of transactions, and the edges are formed by hash links to previous batch transactions. |
8. Transaction management | Like Sawtooth, DGT uses a mechanism for separating transaction families through pluggable transaction processors. This feature makes it possible to use different types of transactions within one stored Ledger mechanism. Additionally, DGT uses a family of topological transactions to form the network topology. | |
9. Tokenization | Missing. | The tokenization mechanism implies the capabilities of system participants to issue their value equivalents for information exchange. |
10. Connecting clients | In both Sawtooth and DGT, the client that defines the application logic (with which the end user is working) connects to the selected node through an API server, that provides all the necessary data. The DGT API server is significantly expanded compared to Sawtooth, which allows for the connection of a system of statistics and general network analysis (Dashboards). | |
11. Byzantine Fault Tolerance | A distinctive feature of Sawtooth among other Hyperledger projects is the embedded resistance to Byzantine attacks: node spoofing, double spending, and other violations of integrity in a distributed network. An additional feature of DGT is the built-in BFT testing system with an attack emulator. | |
12. License | CORE edition – Apache 2.0 GARANASKA edition – AGPL 3.0 |
Backward compatibility with Sawtooth provides the following benefits:
Inheritance of significant functionality that has been security audited and verified in terms of performance.
The use of a consistent technology stack allows DGT to focus on developing distinctive features that most closely match the target characteristics of application scenarios, such as those primarily necessary for establishing digital ecosystems.
Easy integration of such components as Sabre and other modules with high potential.
Last updated
