Magnet: Method-Based Approach Using Graph Neural Network for Microservices Identification
Bibliographic record
Abstract
Monolithic software systems face significant challenges in terms of maintenance, scalability, and portability. To address these challenges, many companies are embracing the microservices architectural style as a more flexible alternative to their monoliths. Microservices structure systems into modular, independent components, enabling easier development, deployment, and maintenance. However, the migration from a monolith to microservices is challenging due to the laborious task of manually identifying and decomposing a system into microservices. Several earlier studies focused on developing approaches to facilitate the migration process. However, the reliance on domain experts to define various parameters and thresholds restricted their use. In this paper, we introduce Magnet, a fully automated microservice identification approach, based on graph neural networks (GNNs). Magnet integrates a GNN model with a fine-grained method-based graph enriched with semantic and static features of the system. It enables accurate microservices identification while simultaneously promoting microservice cohesion and reducing microservice coupling. To validate the accuracy of Magnet, we performed extensive experiments using a set of open-source systems. Quantitatively, we use a set of quality metrics to assess the resulting microservices quality. We also compare our results to established ground truths. Empirical evidence suggests that our fully-automated approach Magnet achieves precision and recall rates of 56% and 68%. Qualitatively, we assess the modularity and functional independence of the resulting microservices by examining their relationships and semantic integrity. This evaluation demonstrates that our fully automated approach yields promising results, underlining its effectiveness in creating modular and coherent microservices.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.003 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.004 | 0.002 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.002 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".