Transformation of data from legacy architecture to updated architecture

A computer-implemented method, comprising:

• receiving, by a determination component of a data transformation system, a request identifying original operational data associated with a first architecture comprising a monolithic application and an intended second architecture, comprising a non-monolithic architecture including a plurality of microservices;
• extracting, by a transformation component, one or more data elements from the original operational data using at least one of rule-based, query-language-based extraction, or language-model-based extraction techniques;
• resolving, by the transformation component, the extracted data elements by correlating the extracted data elements to aspects of the second architecture based at least in part on source code analysis of the second architecture to generate a syntax tree and a logging tree;
• disentangling, by the transformation component, at least a first data element from a second data element of the original operational data using at least one of temporal separation, spatial separation, or language-model-based separation;
• clustering, by the transformation component using a transformation machine learning model, the resolved and disentangled data elements into clusters corresponding respectively to different microservices of the second architecture;
• finalizing transformation of the clustered data elements by duplicating, copying, or transferring at least one clustered data element to a respective microservice to generate updated operational data formatted for use by the second architecture;
• receiving, by a feedback component, entity feedback comprising at least one of match corrections, match additions, or match clarifications relating to the clustering or transformation;
• and
• retraining, by a training component, one or more parameters of the transformation machine learning model based on (i) comparison data between the original operational data and the updated operational data and (ii) the received entity feedback.

A computer program product facilitating transformation of original operational data into updated operational data, the computer program product comprising a non-transitory computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to:

• receive, by a determination component, a request identifying (i) original operational data associated with a first architecture comprising a monolithic application and (ii) an intended second architecture comprising a non-monolithic architecture including a plurality of microservices;
• generate at least one extraction model or language model for processing the original operational data;
• extract, by a transformation component, one or more data elements from the original operational data using at least one of rule-based extraction, query-language-based extraction, or language-model-based extraction techniques;
• resolve the extracted data elements by correlating the extracted data elements to aspects of the second architecture based at least in part on source code analysis of the second architecture to generate a syntax tree and a logging tree;
• disentangle at least a first data element from a second data element of the original operational data using at least one of temporal separation, spatial separation, or language-model-based separation;
• cluster, using a transformation machine learning model, the resolved and disentangled data elements into clusters corresponding respectively to different microservices of the second architecture;
• duplicate or transfer at least one clustered data element to more than one microservice where correlation to plural aspects of the second architecture is established, thereby generating updated operational data formatted for use by the respective microservices;
• and
• retrain one or more parameters of the transformation machine learning model based on (i) comparison data between the original operational data and the updated operational data and (ii) received entity feedback comprising at least one of match corrections, match additions, or match clarifications.