Distributed Matching Engine Feature Modules
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Distributed computing systems, particularly electronic trading platforms, face challenges in handling high-throughput, low-latency, fault-tolerant, and deterministic processing of complex data transaction requests while maintaining system maintainability and flexibility.
Innovation Solution
The implementation of a distributed computing system with feature modules that manage specific types of data transaction requests, such as discretion, pegging, and reserve orders, and a matching engine that processes these requests efficiently, allowing for the apportionment of processing logic and load between feature modules and the matching engine, enhancing system performance and maintainability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If distributed computing systems process complex data transaction requests with high throughput, then processing capacity increases, but system complexity and difficulty of maintaining determinism increase
Solution Approach 1:
The system segments processing logic into distinct feature modules (discretion module, pegging module, reserve module) that each handle specific order types. This modular architecture allows the matching engine to process multiple order types simultaneously through parallel module execution, increasing throughput while maintaining manageable complexity through clear separation of concerns.
Solution Approach 2:
The patent introduces an order management system as an intermediary layer between the matching engine and the exchange. This intermediary handles complex order management tasks including parent request creation, child order generation, and coordination between different feature modules, thereby shielding the core matching engine from complexity while maintaining high processing capacity.
2Adaptability or versatility
If the system handles more types of orders with complex behaviors, then adaptability increases, but system maintainability decreases
Solution Approach 1:
Each order type (discretion, pegging, reserve) is handled by a dedicated feature module with encapsulated logic. This segmentation allows individual modules to be developed, tested, and maintained independently, improving maintainability while supporting diverse order types. New order types can be added by creating new modules without affecting existing functionality.
Solution Approach 2:
The matching engine provides universal processing capabilities that work across all order types through a common interface. The engine can process parent requests, child orders, and various order modifications uniformly, while specific behaviors are determined by the appropriate feature module, thereby maintaining simplicity in the core engine while supporting versatility.
3Device complexity
If processing logic is centralized in one matching engine, then system simplicity increases, but processing speed and throughput decrease
Solution Approach 1:
The system segments the matching engine into multiple specialized feature modules that can execute in parallel. Each module (discretion, pegging, reserve) processes specific order types independently, enabling concurrent processing of multiple orders simultaneously, thereby increasing throughput while maintaining logical organization.
Solution Approach 2:
The patent transitions from a single-threaded sequential processing model to a multi-dimensional parallel processing architecture. The matching engine processes multiple parent requests and child orders across different dimensions (different order types, different feature modules) simultaneously, dramatically increasing processing speed while maintaining system coherence through centralized coordination.
4Loss of time
If the system processes real-time data with time-bound constraints, then responsiveness improves, but system reliability under load decreases
Solution Approach 1:
The system performs preliminary actions by pre-validating order parameters, pre-allocating resources, and pre-establishing processing pipelines for different order types. Parent requests are validated and prepared before being passed to child order generation, reducing critical path latency while maintaining reliability through early error detection and prevention.
Solution Approach 2:
The order management system acts as a mediator that buffers and coordinates real-time processing between incoming orders and the matching engine. This intermediary layer provides fault isolation, ensuring that failures in one feature module or processing stream do not propagate to others, thereby maintaining system reliability under high load while preserving real-time responsiveness.
Data Source
AI summary
In some embodiments a distributed computing system is provided that includes a plurality of different feature modules and a matching engine. The different feature modules each provide different processing for handling parent requests and submitting, to the matching engine, commands for child data transaction requests that are associated with the parent request.