Design and Implementation of Distributed Price Oracle Machine System

To address the single point of failure issue, some Oracle Machine services have adopted a distributed design. Taking the service that provides the Bitcoin to USD price as an example, it aggregates 31 independent price Oracle Machines to provide data to users.

The design of this aggregator allows multiple off-chain Oracle Machines to provide price data by invoking specific methods in response to user requests. These Oracle Machines are usually some externally owned accounts (EOA), and they can provide data not only for the Bitcoin/USD price but also possibly for other currency pairs such as Ethereum/USD.

On-chain contracts perform a series of validation steps when processing data provided by the Oracle Machine:

1. First, read the current contract status and perform a preliminary check.
2. Carry out the necessary preparatory work.
3. Verify each signature data to ensure data integrity and reliable source.
4. Check the sorting of the observed values and select the median as the final result.

Finally, the contract will perform further verification on the obtained results, such as comparing with other price sources to ensure that the deviation is within an acceptable range.

Some platforms also offer more advanced aggregation services, such as Feed Registry. This service aggregates multiple price Oracle Machines, allowing users to more conveniently access price data for different tokens without having to set up each price source individually.

In terms of the pricing mechanism, the Oracle Machine network usually adopts a multi-layer data aggregation:

1. Data Source Aggregation: Collecting raw price data from major trading platforms.
2. Node operator aggregation: Each node obtains and processes data from multiple data sources.
3. Oracle Machine network aggregation: The data from multiple nodes is aggregated again to obtain the final result.

Price updates are usually triggered by two factors: deviation threshold and heartbeat threshold. Different price Oracle Machines may have different parameter settings.

Although this design offers high security and convenience, there are also some limitations. Due to the characteristics of the price update mechanism, data updates may be relatively slow, ranging from a few minutes to a few hours. Therefore, such Oracle Machine services are mainly suitable for application scenarios where real-time price requirements are not high.