Adaptable Replica Server System for Latency and Reliability Trade-offs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing replica server systems are inflexible, as they are typically configured either as synchronous or asynchronous, which does not accommodate varying transaction types and high-speed requirements in financial applications like program/algorithmic trading, where sub-millisecond response times are necessary.
Innovation Solution
An adaptable replica server system that can be configured to manage input data differently based on attributes, allowing it to operate as either synchronous or asynchronous depending on user instructions, with input data being classified as 'safe' or 'fast' to optimize latency and reliability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If synchronous replication is used, then data integrity and reliability are improved, but latency increases
Solution Approach 1:
The patent segments transactions into different categories based on their attributes (e.g., critical vs. non-critical, high-value vs. low-value). Different replication policies are then applied to different segments, allowing synchronous replication for critical transactions requiring data integrity and asynchronous replication for non-critical transactions where latency is less concerning. This resolves the contradiction by not applying a uniform replication policy to all transactions.
Solution Approach 2:
The patent implements local quality by assigning different replication characteristics to different transaction types. Each transaction is evaluated based on its attributes, and the replication behavior is customized accordingly - some transactions receive synchronous replication guarantees while others receive asynchronous replication with best-effort delivery. This localized differentiation allows the system to optimize both reliability and latency depending on transaction-specific requirements.
2Loss of time
If asynchronous replication is used, then latency is reduced, but data integrity and reliability worsen
Solution Approach 1:
The patent divides transactions into segments based on their importance and attributes. Critical transactions with high data integrity requirements are treated differently from non-critical transactions. By segmenting this way, the system can use asynchronous replication for low-latency non-critical transactions while maintaining synchronous replication for critical transactions, thus resolving the contradiction between latency and reliability.
Solution Approach 2:
The patent applies local quality by customizing replication behavior according to transaction attributes. Transactions are evaluated individually or in groups based on their characteristics (value, criticality, timing requirements), and appropriate replication policies are applied locally to each segment. This allows the system to achieve low latency for appropriate transactions while maintaining data integrity for others.
3Device complexity
If a fixed synchronous or asynchronous replication policy is used, then system simplicity is maintained, but adaptability and versatility worsen
Solution Approach 1:
The patent introduces dynamics by making the replication policy configurable and adaptable rather than fixed. The system can dynamically adjust its behavior based on transaction attributes, allowing it to switch between synchronous and asynchronous replication modes depending on the specific transaction requirements. This dynamic configuration capability provides versatility while maintaining reasonable system simplicity through automated policy selection based on transaction characteristics.
Solution Approach 2:
The patent applies parameter changes by allowing the replication policy parameters (synchronous vs. asynchronous, consistency levels, timing thresholds) to be configured based on transaction attributes. Instead of a fixed policy, the system adjusts key parameters dynamically according to the specific needs of each transaction type, thereby achieving adaptability and versatility without excessive complexity.
4Device complexity
If uniform treatment of all transactions is used, then system complexity is reduced, but manufacturing precision and processing accuracy worsen
Solution Approach 1:
The patent segments transactions into different categories based on their attributes (value, criticality, timing sensitivity). By segmenting transactions this way, the system can apply appropriate processing accuracy and replication policies to each segment rather than treating all transactions uniformly. This segmentation enables precise processing for high-value or critical transactions while maintaining simpler handling for less critical ones.
Solution Approach 2:
The patent implements local quality by customizing processing and replication behavior according to transaction-specific attributes. Each transaction segment receives tailored processing accuracy and replication characteristics appropriate to its importance and requirements. This localized differentiation improves processing accuracy for critical transactions without unnecessarily increasing system complexity for all transactions.
Data Source
AI summary
An adaptable replica server system comprising a primary replica being associated with a primary storage for storing information and a primary processing means, and a secondary replica being associated with a secondary storage and a secondary processing means for creating confirmation data upon receiving input data from the primary replica. The replica server system is configurable to work as an asynchronous server replica system and/or as a synchronous server replica system depending on an instruction.