How Distributed Oracle Machines Ensure the Reliability of Price Data

Design and Implementation of a Distributed Price Oracle Machine System ###

In order to address the single point of failure issue, a well-known Oracle Machine platform has adopted a distributed price Oracle Machine design scheme. Taking the BTC to USD price service as an example, the system aggregates 31 independent price Oracle Machines to provide users with reliable price data.

The core of this aggregator design lies in its smart contract. By invoking the transmitters method within the contract, we can view all the off-chain Oracle Machines included in this aggregator. Each off-chain Oracle Machine is an externally owned account (EOA), which can not only provide price data for the BTC/USD aggregator but may also serve other currency pairs (such as ETH/USD) simultaneously.

When users request price data, the off-chain Oracle Machine responds to the requests in the aggregator by calling the transmit method. This process involves several steps:

1. First, the system will read the current contract status and perform a series of security checks.

2. After passing the security check, the system begins the necessary preparation work.

3. Next, verify each signature data to ensure the validity and uniqueness of the signature, while checking the signer's role permissions.

4. Finally, the system will sort the observations, select the median as the final result, and ensure that this result does not exceed the preset threshold.

To further enhance the reliability of the data, the system will also perform additional validation on the final results. For example, in certain application scenarios, the results will be compared with data from other independent Oracle Machine systems to ensure that the price deviation is within an acceptable range.

Oracle Machine registration application

To simplify the access process and reduce governance costs, the platform has introduced the concept of an Oracle Machine registry (Feed Registry). This is a more advanced aggregator that integrates multiple price Oracle Machines. After using the registry, developers no longer need to manually set the price Oracle Machine for each token and can directly read the required price data from the registry.

Multi-level Data Aggregation Mechanism

The price data of the platform is ensured for accuracy and reliability through three levels of aggregation:

1. Data source aggregation: Collecting raw price data from major centralized and decentralized trading platforms.

2. Node operator aggregation: Each node operator obtains price data from multiple independent data aggregation service providers and performs median calculations.

3. Oracle Machine Network Aggregation: Aggregation at the entire network level, typically using the median method, triggered when the number of response nodes reaches a preset threshold.

It is worth noting that not every data update is immediately synchronized to the blockchain. The system has set two trigger parameters: deviation threshold and heartbeat threshold. Only when one of them is met will there be an on-chain update. While this mechanism ensures data stability, it also leads to a relatively slow price update speed, which can range from a few minutes to 24 hours. Therefore, this Oracle Machine system is more suitable for application scenarios where real-time price requirements are not high.