Sequencer Distributes Order Messages to Parallel Engines
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current computerized exchange systems experience high latency and reduced throughput due to the central server's complex processing of trade orders, which requires repetitive processing across multiple gateways, leading to increased processing time and inefficiency.
Innovation Solution
A method and system that utilize a sequencer to sequence order entry messages with status data and sequence numbers, then distribute them to multiple order book engines for parallel processing, eliminating the need for a central server to perform demanding operations and reducing redundant data processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a central server performs complex order matching operations, then order processing can be centralized and controlled, but latency increases and throughput decreases due to repetitive processing across multiple gateways
Solution Approach 1:
The system divides the centralized order processing function into multiple distributed order book engines, each capable of independently processing orders. The sequencer segments the coordination function by assigning sequence numbers to orders and distributing them to appropriate engines, eliminating the single-point bottleneck while maintaining processing reliability through distributed parallel execution
Solution Approach 2:
The sequencer acts as an intermediary between order entry and the distributed order book engines. It receives orders, assigns sequence numbers to ensure proper execution order, and distributes orders to appropriate engines without requiring complex centralized matching logic, thereby reducing latency while maintaining control
2Device complexity
If a central server processes all order entry messages sequentially, then processing logic is simplified, but throughput is reduced due to lack of parallel processing
Solution Approach 1:
The system segments the order processing workload across multiple order book engines that operate in parallel. Each engine maintains simplified processing logic for handling orders within its responsibility, while the sequencer coordinates distribution. This segmentation enables high throughput through parallel execution without requiring each component to handle complex centralized matching logic
Solution Approach 2:
The system transitions from sequential single-threaded processing to parallel multi-threaded processing by distributing orders across multiple engines. The sequencer adds a coordination dimension by assigning sequence numbers and routing orders appropriately, enabling the system to process orders in parallel while maintaining logical sequential integrity
3Adaptability or versatility
If multiple gateways reprocess order entry messages, then each gateway can perform local processing, but redundant processing increases latency and reduces efficiency
Solution Approach 1:
The system extracts the complex order matching and processing logic from multiple gateways and consolidates it in the distributed order book engines. Gateways are reduced to simple message forwarding components that receive orders and forward them to the sequencer, which then distributes to appropriate engines. This extraction eliminates redundant processing while preserving gateway adaptability for local message handling
Solution Approach 2:
Instead of having gateways perform complex processing and forward results, the system inverts the architecture so that gateways simply forward messages and the distributed order book engines perform the processing. This inversion eliminates the redundancy of multiple gateways each attempting to process orders, while maintaining the versatility of the system through distributed engine capabilities
Data Source
AI summary
Methods, apparatus, and systems for processing data transactions use a sequencer communicating with a plurality of data processors and provide improved performance and lower latency.