Lessons learned from hyper-parameter tuning for microservice candidate identification

TL;DR

This paper investigates the impact of hyper-parameter tuning on microservice partitioning, demonstrating significant improvements and highlighting the need to identify optimal optimizers for different scenarios.

Contribution

It reveals the importance of hyper-parameter optimization in microservice identification and provides a reproducible framework for future research in this area.

Findings

Hyper-parameter tuning significantly improves microservice partitioning.

Different optimizers may perform variably across problems.

Open source tools facilitate reproducibility and further exploration.

Abstract

When optimizing software for the cloud, monolithic applications need to be partitioned into many smaller *microservices*. While many tools have been proposed for this task, we warn that the evaluation of those approaches has been incomplete; e.g. minimal prior exploration of hyperparameter optimization. Using a set of open source Java EE applications, we show here that (a) such optimization can significantly improve microservice partitioning; and that (b) an open issue for future work is how to find which optimizer works best for different problems. To facilitate that future work, see [https://github.com/yrahul3910/ase-tuned-mono2micro](https://github.com/yrahul3910/ase-tuned-mono2micro) for a reproduction package for this research.