Аннотация
This systematic review examines how higher-education institutions used virtualisation between 2018 and April 2025, with particular attention to its implications for educational quality, learning design, and operational efficiency. Searches across six scholarly databases identified ten quasi-experimental studies from eight countries and six academic fields, reflecting diverse institutional and socio-economic contexts. The evidence shows that a three-layer tool chain combining virtual desktops, container-orchestrated cloud labs, and immersive front ends can sustain or improve learning outcomes while reducing on-site hardware needs and shortening laboratory time, primarily by increasing time-on-task, supporting experiential learning, and reducing setup-related cognitive load. Where, investigators logged server telemetry, central-processing-unit load stayed below 4%, and memory use remained under 2 GB even as enrollment doubled. Learning gains reached large effect sizes in networking, biology microscopy, and engineering tasks delivered through augmented reality, although these effects were often observed in small cohorts and quasi-experimental settings. However, most studies omitted capital and operating cost data, relied on modest sample sizes, and lacked random assignment or assessor blinding, which limits causal inference and economic generalisation. Overall, the findings suggest that virtualisation can enhance learning effectiveness and scalability when pedagogically integrated, but the current evidence base remains methodologically heterogeneous. Future research should adopt common performance and cost metrics, enroll larger cohorts, and report longitudinal outcomes to guide evidence-based investment in campus-wide virtualisation.