Shared-Memory Failover Architecture for Low-Latency Trading

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing electronic trading systems face challenges in achieving fault tolerance with minimal recovery time and performance degradation due to geographic separation of backup systems, leading to communication latency and potential data loss during failover scenarios.

Innovation Solution

Integrate core components of the trading system on a single processing platform using shared memory communication and implement a fault tolerance mechanism that synchronizes and switches to a backup instance upon failure, ensuring all components fail over together, minimizing latency and data loss.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If backup systems are located in different geographic regions from the primary instance, then fault tolerance and system availability are improved, but communication latency increases and data loss risk increases during failover

Engineering Contradiction:

Improvefault toleranceVSAvoidcommunication latency

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system segments fault tolerance into two distinct mechanisms: synchronous replication for critical state data (maintained in shared memory within the same geographic location) and asynchronous replication for full backup (maintained in geographically separate locations). This segmentation allows the system to optimize each replication type for its specific purpose, minimizing latency for critical operations while maintaining geographic separation for disaster recovery.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces a shared memory communication infrastructure as an intermediary between the primary instance and backup systems. This shared memory acts as a high-speed buffer that enables rapid state synchronization without the latency penalties of traditional network communication, effectively mediating between the need for geographic separation and the need for low-latency failover.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Reliability

If backup systems are located in different geographic regions from the primary instance, then fault tolerance is improved, but data loss during failover increases

Engineering Contradiction:

Improvefault toleranceVSAvoiddata loss

Core Design Contradiction:

ReliabilityVSLoss of information

Solution Approach 1:

The system performs preliminary synchronization of the backup instance state with the primary instance state before failover is needed. Through continuous synchronous replication of critical state data to shared memory and periodic asynchronous replication of full system state, the backup is preliminarily prepared with up-to-date information, minimizing data loss when failover occurs.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent implements a cushioning mechanism through shared memory that buffers and protects against data loss during failover. The shared memory maintains a synchronized copy of critical state data that can be immediately transferred to the backup instance, cushioning the system against the effects of geographic separation and network latency during critical failover moments.

Inventive Principle:

Principle #11Beforehand cushioning (Prior cushioning)

3Loss of time

If core components are integrated on a single processing platform with shared memory communication, then communication latency is reduced, but system complexity increases

Engineering Contradiction:

Improvecommunication latencyVSAvoidsystem complexity

Core Design Contradiction:

Loss of timeVSDevice complexity

Solution Approach 1:

The patent merges multiple core components (primary instance, backup instance, and shared memory communication infrastructure) onto a single processing platform. This consolidation eliminates the need for external network communication between these components, dramatically reducing communication latency while managing complexity through unified resource management and coordinated failover protocols.

Inventive Principle:

Principle #5Merging (Combining)

4Stability of the object's composition

If synchronous replication is used for backup systems, then data consistency is improved, but communication bandwidth requirements increase

Engineering Contradiction:

Improvedata consistencyVSAvoidcommunication bandwidth

Core Design Contradiction:

Stability of the object's compositionVSLoss of energy

Solution Approach 1:

The patent applies different replication qualities to different data types and timeframes: synchronous replication is used for critical state data that requires immediate consistency (local quality optimization), while asynchronous replication is used for full backup systems where eventual consistency is acceptable. This local quality differentiation reduces overall bandwidth requirements while maintaining data consistency where absolutely necessary.

Inventive Principle:

Principle #3Local quality

Data Source

PatentUS12487894B2Fault tolerant architecture

Publication Date: 2025.12.02 CHICAGO MERCANTILE EXCHANGE INC

AI summary

A system/process provides fault tolerance to an integrated component system which integrates core components of a transaction processing system into a single processing platform, i.e., a single server, enabling elimination of the network interconnects and associated latencies introduced thereby in favor of much faster interconnects, such as inter-process communication and shared memory communication messaging, where a failure of any one component necessitates failing over the entire system to a backup thereof.