Abstract
This paper presents the architecture of the Kazakhstani National Virtual Observatory (KazVO)—an International Virtual Observatory Alliance (IVOA) compliant node that unifies the heterogeneous observational datasets of the Fesenkov Astrophysical Institute (FAI). The architectural foundation of the system is built upon a partitioned Data Lake, a German Astrophysical Virtual Observatory (GAVO) Data Center Helper Suite (DaCHS) publishing backend, and a PostgreSQL database in which metadata for published observational products are mapped to IVOA ObsCore, while catalogs and simulation grids use dedicated relational tables. We follow Open Science principles by introducing a metadata-only showcase model based on a four-class data embargo mechanism. Machine-to-Machine services deployed via IVOA protocols are exposed through the KazVO publishing registry registered in the global Registry of Registries (RofR), enabling analysis of Kazakhstani observational assets within external clients such as Tool for OPerations on Catalogues And Tables (TOPCAT), Aladin, and PyVO. The current implementation combines operational VO publication services and targeted photometric processing with scheduled, batch-mode data ingestion, supporting access to published historical and modern observational holdings. Replication has been validated on pilot datasets, while routine deployment at the observing sites and automated reduction of modern spectroscopic streams remain under development. The architecture provides a framework for progressively integrating these components into a complete data lifecycle.