Client Device Market Data Coalescing for Queue Overflow
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In electronic trading environments, client devices often face issues with processing high volumes of market data updates, leading to queue buildup and outdated information, which can result in disastrous trading decisions or system failures due to the inability to handle frequent price, order, or fill updates.
Innovation Solution
Implementing a system that coalesces market data updates at a client device, where incoming updates are combined and processed as a single update, reducing processor time and ensuring the graphical user interface and other applications receive the most current market information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If the client device processes all incoming market data updates individually, then the system maintains complete information accuracy, but the message queue builds up and causes processing delays
Solution Approach 1:
The patent combines multiple incoming market data updates into a single coalesced update when they occur within a predetermined time window. This merging approach prevents message queue buildup and processing delays while ensuring the GUI receives the most recent market data, thus resolving the contradiction between information accuracy and processing time.
2Productivity
If the client device uses a fast processor to handle high-volume market data updates, then processing speed increases, but other applications like the graphical user interface are prevented from performing operations
Solution Approach 1:
By merging multiple market data updates into a single coalesced update, the system reduces the total number of processing operations required. This allows the processor to handle market data efficiently while leaving sufficient processing capacity for the graphical user interface and other applications to remain responsive.
Solution Approach 2:
The system uses periodic coalescing intervals (predetermined time windows) to batch process market data updates. This periodic approach creates predictable processing cycles that allow the GUI and other applications to perform operations between update batches, maintaining system responsiveness.
3Loss of information
If the internal message queue delivers all messages in order, then complete information is provided to the client, but the queue grows indefinitely until overflow occurs
Solution Approach 1:
The patent merges consecutive market data updates into coalesced updates, significantly reducing the number of messages that need to be queued and delivered. This prevents indefinite queue growth and overflow while ensuring the client receives the most recent market data, thus maintaining both information completeness and system stability.
Solution Approach 2:
The system discards intermediate update messages that are subsumed by coalesced updates, recovering only the essential most-recent market data. This selective discarding prevents queue overflow while maintaining the integrity of the market information delivered to the client.
Data Source
AI summary
A client device coalesces data received from an exchange, and provides a client application such as a graphical user interface with the opportunity to process fewer, but up-to-date, data updates from an exchange when a large volume of prices becomes available. Accordingly, the trader can be assured of receiving updated information that are fed to the client applications at a rate that is cohesive with that client device's processing speed.