STRIDE-DB: A Structured, Risk-Aware and Intelligent Framework for Legacy Database Transformation in Hybrid Cloud Environments

Legacy database systems are still being required to support mission-critical applications, which need to be scalable, interoperable, resilient, governed, cloud-operated and more. The problem with modernizing these systems is that transforming the data from a database is not only hard to do because the data is much more than just schemas and data transfer, but also because of structural dependency, semantic correspondence, data quality, complexity of the transform, migration risk, resources required, deployment constraints, etc. Most of the current solutions address this issue individually, and not as part of a migration decisioning process. In this paper, the authors propose a hybrid cloud-based transformation framework for the evolution of the intelligent legacy database, called as Structured Transformation and Risk-aware Intelligent Database Evolution (STRIDE-DB). The intelligent cloud recommendation, automatic transformation and migration planning, validation tool, explainable AI, machine learning risk prediction, ensemble learning, business rule integrity, data quality, dependency discovery and verification, and metadata discovery are all features of the STRIDE-DB architecture. The following tests (BaseLine, SemanticStress, DataQualityStress, DependencyStress, CombinedStress and optimized migration) are run on a controlled experimental layer. The current implementation was tested with seven source tables with about 59825 records and 38 attributes of a relational database. The knowledge representation layer created 49 nodes and 43 edges, while the graph representation learning created 64-dimensional node embeddings. The pre-migration risk indicator was determined to be 0.1717 (low risk), and the transformation-complexity indicator was determined to be 0.2184 (low complexity). Out of 38 attribute mappings, 35 were chosen to be automatically transformed, and three were kept for further review due to a lack of confidence. The average target attribute coverage was 94.05% while the average target completeness was 99.66%. Furthermore, the framework offers a number of machine learning classifiers, multiple classifiers, ensemble learning, explanation using SHAP, auto-generated SQL and SQL-recommended and explanation for the hybrid cloud. The results show that the concept of integrated pipeline STRIDE-DB is viable for database modernization for structured and risk-aware databases. It is a small table-level data set, however, and is just a feasibility examination of the implementation stage, as a substitute for a generalization skill in the machine-learning evaluation. More large-scale, multi-dataset and multi-scenario evaluations to prove the predictive models are valid are required.

Authors

Institutions

Publication Details

Journal
International Journal of Electronics and Communication Engineering
Published
2026-11-12
DOI
https://doi.org/10.14445/23488549/ijece-v13i10p107
Primary Topic
Advanced Database Systems and Queries
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

STRIDE-DB: A Structured, Risk-Aware and Intelligent Framework for Legacy Database Transformation in Hybrid Cloud Environments

S. Rama Sree, Suneel Pappala, N. Siva Kumar
International Journal of Electronics and Communication Engineering
Advanced Database Systems and Queries
article

STRIDE-DB: A Structured, Risk-Aware and Intelligent Framework for Legacy Database Transformation in Hybrid Cloud Environments

S. Rama Sree, Suneel Pappala, N. Siva Kumar
article en

Abstract

Legacy database systems are still being required to support mission-critical applications, which need to be scalable, interoperable, resilient, governed, cloud-operated and more. The problem with modernizing these systems is that transforming the data from a database is not only hard to do because the data is much more than just schemas and data transfer, but also because of structural dependency, semantic correspondence, data quality, complexity of the transform, migration risk, resources required, deployment constraints, etc. Most of the current solutions address this issue individually, and not as part of a migration decisioning process. In this paper, the authors propose a hybrid cloud-based transformation framework for the evolution of the intelligent legacy database, called as Structured Transformation and Risk-aware Intelligent Database Evolution (STRIDE-DB). The intelligent cloud recommendation, automatic transformation and migration planning, validation tool, explainable AI, machine learning risk prediction, ensemble learning, business rule integrity, data quality, dependency discovery and verification, and metadata discovery are all features of the STRIDE-DB architecture. The following tests (BaseLine, SemanticStress, DataQualityStress, DependencyStress, CombinedStress and optimized migration) are run on a controlled experimental layer. The current implementation was tested with seven source tables with about 59825 records and 38 attributes of a relational database. The knowledge representation layer created 49 nodes and 43 edges, while the graph representation learning created 64-dimensional node embeddings. The pre-migration risk indicator was determined to be 0.1717 (low risk), and the transformation-complexity indicator was determined to be 0.2184 (low complexity). Out of 38 attribute mappings, 35 were chosen to be automatically transformed, and three were kept for further review due to a lack of confidence. The average target attribute coverage was 94.05% while the average target completeness was 99.66%. Furthermore, the framework offers a number of machine learning classifiers, multiple classifiers, ensemble learning, explanation using SHAP, auto-generated SQL and SQL-recommended and explanation for the hybrid cloud. The results show that the concept of integrated pipeline STRIDE-DB is viable for database modernization for structured and risk-aware databases. It is a small table-level data set, however, and is just a feasibility examination of the implementation stage, as a substitute for a generalization skill in the machine-learning evaluation. More large-scale, multi-dataset and multi-scenario evaluations to prove the predictive models are valid are required.

International Journal of Electronics and Communication EngineeringVol. 13(10)
Aditya University (IN)
Industry, innovation and infrastructure
Advanced Database Systems and Queries
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon