From Microservices to AI-Native Systems: Rethinking Enterprise Architecture Through the ANARCH Framework
International Journal on Science and Technology · 14.04.2026 · 17 (2)
DOI: 10.71097/ijsat.v17.i2.10641 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal on Science and Technology, online erschienen am 14.04.2026. Verfasst von Samer Yadav und 1 weiteren Autor:innen.
Abstract (Originalsprache)
Enterprise software architectures have undergone successive transformations - from monolithic systems to service-oriented architectures, and subsequently to microservices. However, the emergence of large-scale artificial intelligence workloads, including large language models, real-time inference pipelines, and autonomous decision systems, has exposed fundamental limitations in microservices-based architectures. Conventional microservices patterns assume stateless, deterministic service interactions and do not inherently accommodate the probabilistic reasoning, continuous learning, high-dimensional data flows, and GPU-accelerated compute requirements that characterize AI-native workloads. This paper introduces the AI-Native Architecture (ANARCH) framework, a structured reference architecture that reconceptualizes enterprise system design around AI as a first-class architectural primitive rather than an auxiliary capability. ANARCH defines six foundational pillars—Cognitive Service Mesh, Adaptive Data Fabric, Model Lifecycle Sovereignty, Inference Elasticity, Autonomous Orchestration, and Continuous Governance - and specifies formal interaction constraints across pillars to prevent architectural fragmentation. Comparative evaluation against microservices-based, our prior LEAIM layered integration model, and embedded AI integration patterns demonstrates that ANARCH achieves superior modularity, inference throughput elasticity, governance depth, and resilience under heterogeneous AI workloads. By positioning AI not as an integration concern but as the organizing principle of enterprise architecture, ANARCH provides a systematic foundation for next-generation AI-native enterprise systems.
Zitieren
APA
Yadav, S., & Mewani, D. (2026). From Microservices to AI-Native Systems: Rethinking Enterprise Architecture Through the ANARCH Framework. International Journal on Science and Technology, 17(2). https://doi.org/10.71097/ijsat.v17.i2.10641
Vancouver
Yadav S, Mewani D. From Microservices to AI-Native Systems: Rethinking Enterprise Architecture Through the ANARCH Framework. International Journal on Science and Technology. 2026;17(2). doi:10.71097/ijsat.v17.i2.10641
BibTeX
@article{Yadav2026,
author = {Yadav, Samer and Mewani, Dheeraj},
title = {From Microservices to AI-Native Systems: Rethinking Enterprise Architecture Through the ANARCH Framework},
journal = {International Journal on Science and Technology},
year = {2026},
volume = {17},
number = {2},
doi = {10.71097/ijsat.v17.i2.10641}
}
EUVIMED