Legacy Data Utility Scoring for Microservice Prioritization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Legacy system modernization faces challenges in utilizing dark data, as approximately 90% of sensor-generated data goes unused, and the cost of conversion or re-platforming is prohibitively expensive, making it difficult to prioritize which candidate microservices to develop for exposing valuable data insights.
Innovation Solution
A method utilizing intra-analysis and meta-analysis to measure the utility of data metrics in legacy systems, correlating them with candidate microservices, and generating insights to prioritize exposure and reduce progressively variant dark data by scoring the useful lifespan of data exposed by each microservice.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If data is exposed through microservices to modernize legacy systems, then data utility and accessibility are improved, but conversion and re-platforming costs increase prohibitively
Solution Approach 1:
The patent segments the legacy system into candidate microservices, analyzing each independently for dark data potential. By dividing the monolithic conversion problem into smaller microservice units, the system can prioritize which segments to convert based on data utility scores, reducing overall conversion costs while maintaining data accessibility.
Solution Approach 2:
The patent changes the parameter of data evaluation from simple exposure to multi-metric analysis including intra-analysis utility, meta-analysis utility, data density, and cosine similarity. This parameter transformation enables cost-effective prioritization by identifying which data metrics provide the highest utility before conversion costs are incurred.
2Productivity
If all candidate microservices are developed to expose data, then data accessibility is improved, but development resources and time are wasted on low-value microservices
Solution Approach 1:
The patent performs preliminary analysis of candidate microservices before development by calculating dark data utility scores using intra-analysis and meta-analysis. This preliminary evaluation identifies which microservices will provide the most value, allowing organizations to prioritize development efforts and avoid wasting time on low-value microservices.
Solution Approach 2:
The patent implements a feedback mechanism where dark data utility scores are calculated and used to guide microservice development priorities. The system continuously evaluates data metrics and provides feedback on which candidate microservices should be developed first, optimizing the allocation of development resources and time.
3Ease of manufacture
If data is retained in legacy systems without exposure, then storage costs are reduced, but data becomes dark and loses value rapidly
Solution Approach 1:
The patent transforms the evaluation of retained data from a static storage cost consideration to a dynamic multi-parameter analysis including intra-analysis utility, meta-analysis utility, and data density. This parameter change reveals which retained data metrics have high value potential, enabling selective exposure that recovers data value while minimizing unnecessary storage costs.
Solution Approach 2:
The patent applies partial action by selectively exposing only those data metrics that demonstrate high dark data utility scores through intra-analysis and meta-analysis. Rather than exposing all retained data or none at all, the system identifies and exposes only the portion of data that provides maximum value, optimizing both storage cost efficiency and data value recovery.
Data Source
AI summary
In an approach for derivation of progressively variant dark data utility for legacy system candidate microservices, a processor analyzes a variability of data stored in a legacy system for a plurality of data metrics. A processor measures, for each data metric, a utility of the data using an intra-analysis and a meta-analysis. A processor correlates the plurality of data metrics to candidate microservices. A processor generates insights for the candidate microservices based on the utility of the data.