Аңдатпа
The paper presents the architecture of a cloud-based software as a service (“SaaS”) computational service designed for numerical modeling using an enhanced fractional‐differential stochastic model. The aim of the study is to create a modular system that enables users to perform high‐precision calculations without the need for their own computational resources. The proposed solution consists of interconnected subsystems for data ingestion, queuing and preprocessing, the mathematical solver, result analysis and visualization, monitoring, logging, and security. Each subsystem is implemented using modern tools (Julia, Python, Django, Prometheus, etc.), communicates via APIs, and relies on a centralized PostgreSQL datastore. The result is a scalable, fault-tolerant architecture that automates request intake, computation, and result interpretation. The conclusion emphasizes that this distributed approach is effective and can be adapted for other scientific modeling and analytics tasks.