Federated Reference Data Management via Standardized Information Messages
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional systems face challenges in managing and processing reference data across multiple domains in a large multi-line financial organization, leading to data complexity, operational risks, and inefficiencies due to reliance on single centralized systems and brittle model representations.
Innovation Solution
Implementing a multi-domainal reference data management module that utilizes a federated system of record approach, enabling real-time data acquisition, storage, and publication through standardized information messages, and integrating with trade processing and back-office applications to support front-to-back office requirements, including straight-through processing and data validation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single centralized system of record is used to manage reference data, then data governance and control are simplified, but the system becomes brittle and suffers from time-to-market issues due to change management constraints
Solution Approach 1:
The patent divides the reference data management system into multiple domain-specific systems of record, each handling specific asset classes or data domains independently. This segmentation allows each system to evolve and adapt without affecting others, resolving the contradiction between centralized control and flexibility.
Solution Approach 2:
The patent introduces a federated architecture dimension that overlays the traditional centralized model, creating a multi-layered system where centralized governance coexists with distributed domain autonomy. This allows simultaneous achievement of both control and adaptability.
2Reliability
If reference data is obtained from multiple external source systems and internal systems, then data completeness and accuracy are improved, but data management complexity and operational risk increase
Solution Approach 1:
The patent introduces a federated reference data management layer that acts as an intermediary between multiple data sources and consuming systems. This intermediary standardizes data integration, manages complexity centrally, while allowing distributed data sourcing, thus improving reliability without proportionally increasing management complexity.
Solution Approach 2:
The patent creates a universal federated management framework that handles multiple data sources, validation rules, and distribution channels through a single standardized interface, reducing the operational complexity that would otherwise arise from managing diverse data sources.
3Device complexity
If a single physical model is used for all reference data within a domain, then system simplicity is maintained, but the model becomes brittle and cannot accommodate diverse asset classes
Solution Approach 1:
The patent segments the reference data model into domain-specific models tailored to different asset classes (e.g., equities, fixed income, derivatives). Each domain model is optimized for its specific requirements, improving representation accuracy without requiring a single overly complex universal model.
Solution Approach 2:
The patent applies local quality by allowing different data models and structures in different domains based on their specific requirements. Each asset class receives a customized model structure that fits its characteristics, rather than forcing a one-size-fits-all approach.
4Reliability
If reference data management follows traditional change management processes, then data integrity and control are maintained, but time-to-market for new data is excessive
Solution Approach 1:
The patent implements dynamic change management where the rigor of governance processes adapts based on the domain and change type. Critical data changes undergo strict review, while routine updates in stable domains can be deployed more rapidly, optimizing both control and speed.
Solution Approach 2:
The patent enables preliminary data modeling and validation in development environments before production deployment. This allows extensive testing and preparation upfront, reducing the need for lengthy change management cycles in production and accelerating time-to-market.
Data Source
AI summary
A system and method for multimodal contribution and management of multi-domainal reference data is disclosed. A processor executes an application that utilizes distribution and lookup services on an authoritative data cache to confirm that a reference data record or attribute does not exist and creates a message compliant with a standardized information message (SIM) format; and inspects the contents of the SIM to determine the domain of the data and transmits the message to a corresponding domainal reference data system of record among a plurality of reference data systems within a federated reference data system. A receiving domain system acts as a system of record to one or more reference data domain classes related to the reference data; and initiates the processes of storing, versioning, and publication of the reference data for the domain of reference data for which the reference data system is being designated a system of record.